R. David G. Pyne (Author) - Groundwater Recharge and Wells - A Guide To Aquifer Storage Recovery-CRC Press (1995)
R. David G. Pyne (Author) - Groundwater Recharge and Wells - A Guide To Aquifer Storage Recovery-CRC Press (1995)
R. David G. Pyne (Author) - Groundwater Recharge and Wells - A Guide To Aquifer Storage Recovery-CRC Press (1995)
RECHARGE
and WELLS
A Guide to
Aquifer Storage Recovery
R. David G. Pyne
LEWIS PUBLISHERS
A CRC Press Company
Boca Raton London New York Washington, D.C.
Library of Congress Cataloging-in-Publication Data
Pyne, R. David G.
Groundwater recharge and wells a guide to aquifer storage
recovery I R. David G. Pyne.
p. em.
Includes bibliographical references and index.
ISBN 1-56670-097-3
1. Aquifer storage recovery. 2. Title.
TD404.5.P95 1994
627'.56--dc20 94-27575
CIP
This book contains information obtained from authentic and highly regarded sources. Reprinted material
is quoted with permission, and sources are indicated. A wide variety of references are listed. Reasonable
efforts have been made to publish reliable data and information, but the authors and the publisher cannot
assume responsibility for the validity of all materials or for the consequences of their use.
Neither this book nor any part may be reproduced or transmitted in any form or by any means, electronic
or mechanical, including photocopying, microfilming, and recording, or by any information storage or
retrieval system, without prior permission in writing from the publisher.
AThe consent of CRC Press LLC does not extend to copying for general distribution, for promotion, for
creating new works, or for resale. Specific permission must be obtained in writing from CRC Press LLC
for such copying.
Direct all inquiries to CRC Press LLC, 2000 N.W. Corporate Blvd., Boca Raton, Florida 33431.
Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are
used only for identification and explanation, without intent to infringe.
Water is power. Control of water is, therefore, the currency of personal, regional
and national ambitions. Existing dams, surface reservoirs, major pipelines and
pumping stations continue to serve us well, and appeal to that side of us which
needs a monument to our achievements. The days are numbered, however, for
many major water resources development projects that are in the planning stages.
Changing priorities are causing us to reconsider the wisdom of traditional engi-
neering solutions that were quite acceptable just a few years ago. More sophis-
ticated, but less visible monuments are needed. The time has come to take a harder
look at storing water below ground in reservoirs provided by nature.
Groundwater levels continue to decline around the world in response to in-
creasing withdrawals to meet the needs of an expanding population. At the same
time, surface waters are proving increasingly unreliable to meet these growing
needs, despite numerous major and costly programs to store and divert water from
a diminishing number of uncontrolled rivers. Intensive urban and agricultural
development is draining our lands during times of rainfall, and depleting our
groundwater resources at an alarming rate during times of drought. As the global
nature of this challenge becomes more clear, we are slowly coming to grips with
the need for sustainable water management.
Adequate storage is the key to sustainable water management. We have suffi-
cient water in most cases, however we have difficulty storing it when it is
available so that it will also be available when and where we need it. Storage in
surface reservoirs is expensive and increasingly perceived as an unacceptable
exchange for valued ecosystems. Storage below ground has been limited for a
variety of reasons. Among these, the principal constraints have been technical,
although political, legal and other constraints have proved significant.
In recent years, development of technology for artificial recharge of aquifers
has accelerated. Most of the technical constraints have been addressed and re-
solved through research and experience at many sites. Systems to recharge aqui-
fers through surface methods such as basins and in-channel structures are func-
tioning reasonably well, however their widespread application is frequently lim-
ited by hydrogeologic constraints and the availability of land at reasonable cost.
Many areas would benefit from aquifer recharge but can do so only through wells.
While many test programs have been conducted, there are few operational well
recharge systems around the world.
Recent technical advances and operational experience have demonstrated that
well recharge is feasible and cost-effective. While many papers have been written
regarding specific recharge projects, the author is aware of only a few papers and
books that assemble the technology. This book provides a guide to those who
would endeavor to successfully recharge aquifers through wells.
A key element of this technology is the need to control well plugging due to
suspended solids in the recharge water, bacterial activity, gases and other causes.
Experience has demonstrated that dual-purpose wells, equipped for both recharge
and recovery, are best able to achieve recharge objectives while controlling
plugging. These dual-purpose wells are called "aquifer storage recovery," or ASR
wells. They are designed and operated differently than normal production wells
or injection wells. This book presents ASR technology, as it has evolved over the
past 25 years in the United States.
During the past six years, many interested in the field of groundwater recharge
have urged the writing of this book, so that they and others can more easily grasp
the ASR vision and implement this technology to meet their various needs. As
time has gone by and new technical issues have been met and resolved, the body
of knowledge that comprises ASR technology and experience has evolved and
matured. With 20 ASR systems now operational in the United States and about 40
more in various stages of development, it is appropriate to distill the variety of
technical and other experiences into this book, a guide to aquifer storage recovery.
Following the procedures suggested in these chapters should enhance the likeli-
hood of achieving success with groundwater recharge through wells.
Although ASR is not "high tech," neither is it "low tech."Understanding the
issues that have been encountered at other sites, and the steps that have lead to
successful resolution of these issues, can provide great help to those considering,
planning or implementing new groundwater recharge projects.
ASR is a new, efficient and cost-effective tool for water resources manage-
ment. Although developed primarily within the United States, it builds upon prior
experience, primarily in The Netherlands and Israel, relating to artificial recharge
of groundwater through wells. ASR is therefore equally applicable in other
countries, many of which have severe water supply challenges.
It is hoped that, by presenting the ASR technology and demonstrating its
feasibility and cost-effectiveness to meet global needs for sustainable water
development, this book will help to defuse political tensions, improve human
welfare and enhance the reliable supply of good quality water at reasonable cost
to people around the world.
R. David G. Pyne
ABOUT THE AUTHOR
The author's experience with aquifer storage recovery has evolved from the
glimpse of an idea in 1976, to a grand vision during the mid-1980s, and to the
widespread and rapidly growing implementation of this vision as of 1994. The
experiences and understanding of many people have guided the development of
ASR technology during this period. Some have assisted directly with the writing
of this book, while many others have played important roles during the develop-
ment and refinement of ASR technology at many locations.
CH2M HILL Inc., consulting engineers, has provided a strong technical plat-
form, without which it is unlikely that ASR would have developed as successfully
and as rapidly as it has. By providing ready accessibility to technical experts in
a wide variety of fields, it has been possible to address and rapidly resolve many
issues that have arisen along the way, both technical and otherwise. By providing
a broad network of office locations and client needs, throughout the United States
and also overseas, we have been able to continually redefine ASR technology to
meet a rapidly growing number and variety of applications, thereby deepening our
understanding of the technology and how it may best be applied to meet these
needs.
Through their hard work and creative ideas, many of my friends at CH2M
HILL have earned recognition for their valuable contributions to the development
of ASR technology. Richard P. Glanzman, Denver, Colorado, has provided con-
tinuous enthusiastic encouragement and sage guidance on geochemistry for vir-
tually all ASR projects. Chapter 5, Geochemistry, is his contribution, for which
he has achieved widespread recognition. Some of the words are mine but many
are his, and the deep understanding upon which they are all based is his.
Kevin Bral, Denver, Colorado, has similarly contributed greatly to the techni-
cal success of many ASR systems and has earned recognition for his efforts in
many areas, particularly including well and wellhead design, test procedures and
the development of the column testing equipment and downhole control valve.
His contributions to Chapter 3, Design of ASR Systems, are greatly appreciated.
Richard Randall, Phoenix, Arizona, has worked for many years to gain improved
understanding of well clogging issues. The current status of his work is included
in Chapter 4, Selected ASR Technical Issues. Margaret !bison, Reston, Virginia,
has contributed her special knowledge of coring, core handling and analysis
procedures along with her enthusiasm and boundless perseverance. Ken McGill,
Philadelphia, Pennsylvania, has persevered through a daunting sequence of tech-
nical challenges on a most complex project at Swimming River, New Jersey,
emerging successfully and contributing valuable experience to guide many other
projects with similar challenges. Dan Wendell, Los Angeles, has contributed his
understanding that comes from having helped to develop and operate the early
ASR system at Goleta, California. His insightful questions, suggestions and
observations have added much. Also in Los Angeles, Terry Foreman has guided
the local development of ASR to meet the needs of water users in southern
California, contributing greatly to the broadening applications of this technology.
Albert Muniz, Deerfield Beach, Florida, has pushed back the technical and
regulatory frontiers through his energetic and capable direction of several chal-
lenging ASR projects in Florida. Modelling of ASR systems has benefitted from
the strong capabilities of John Glass, Peter Kwiatkowski, AI Aikens, Margaret
O'Hare and Sumant Gupta. Their valuable work is reflected in these pages.
Through their efforts on several ASR projects, many others within CH2M
HILL have helped to move this technology forward, among whom are Mark
Lucas, Clifford Bell, Doug Dronfield, Tom Buchanan, Greg Tate, Bryan McDonald,
Ross Sproul, Robert Peterson, Dr. Jose Ignacio Garcia-Bengochea, Bart Zeigler,
Sean Skehan, Tim Sharp, Phil Waller, Mark McNeal, Mike Micheau, Paul Wallace,
Rene Brewster, John McLeod, Paul Thornhill, Larry Amans, Courtney Hemenway,
Andrea Aikin, Peter Livingston, Derrik Williams, Fritz Carlson, Jeff Barry, Stuart
Brown, Pamela Brody, David Livise and Ken Trotman. Many others are now
sharing in the ASR vision and will build upon the foundations laid by those who
have gone before.
Outside CH2M HILL, many have contributed through their suggestions, sup-
port and encouragement. Through his early grasp of the ASR vision and his
continuing attention to details as our shared understanding of the ASR technology
evolved, Sam Stone at the Peace River Water Treatment Plant, Florida, has made
a valuable and enduring contribution to the field. The American Water Works
Association Research Foundation (A WWARF) deserves much appreciation for
funding the research into reduction of disinfection byproducts that has been
observed to occur during ASR storage, as presented in Chapter 4. The Las Vegas
Valley Water District, Nevada, and Thames Water Utilities, England, also shared
in funding this important work. The Southwest Florida Water Management Dis-
trict deserves great credit for funding the initial ASR test program at Lake
Manatee in 1979, at a time when conventional wisdom was that the project would
undoubtedly fail. Without their early financial support, ASR might still be a
concept instead of a reality. Similarly, the South Florida Water Management
District has grasped the potential value of ASR technology to a region with
seasonally available supplies and many water resource challenges. Their early
financial support for the Marathon ASR program in the Florida Keys has contrib-
uted greatly to definition of potential ASR applications in brackish and seawater
aquifers.
Writing this book has required a lot of work, made easier through the efforts
of Tammy Nelson, Mark and Joyce Bradley, Michelle Colletti and Gary Mardock,
who assisted with the word processing, editing and graphics. Mr Hang Soo-Hoo
assisted with library support, unearthing sources from many strange places. I am
also grateful to Brian Lewis, whose early and continuing encouragement regard-
ing the writing and publishing of this book provided a timely stimulus. And for
her skill in editing the manuscript and preparing it for printing, and also for her
humor, Julie Spadaro has earned my gratitude for a job well done.
Most importantly, my family has waited patiently for too long for this book to
be finished so that we can get on with our lives. To Emily, thank you for your
support and encouragement. To Christopher, yes, it is finally, really, 100%
finished.
CONTENTS
Preface ................................................................................................................. iv
Acknowledgments .............................................................................................. vii
List of Figures .................................................................................................... xiv
Chapter 1
Introduction ........................................................................................................ 1
1.1 Water Management: A Global Challenge ............................................ !
1.2 ASR: A New Water Management Too1 .............................................. .4
1.3 What is ASR? ....................................................................................... 6
1.4 Historical Development of ASR .......................................................... 9
1.5 ASR Applications to Meet Water Management Needs ..................... 18
1.6 Objectives of This Book .................................................................... 21
Chapter 2
ASR Program Development ............................................................................ 23
2.1 Introduction ........................................................................................ 23
2.2 Phase 1: Feasibility Assessment and Conceptual Design .................. 25
Recharge Objectives .......................................................................... 26
Water Supply ..................................................................................... 27
Recharge Water Quality ..................................................................... 28
Water Demand ................................................................................... 30
Hydrogeology .................................................................................... 32
Selection of Recharge Processes ........................................................ 34
Site Selection ..................................................................................... 34
Conceptual Design ............................................................................. 35
Existing Well vs. New Well .............................................................. 36
Hydrogeologic Simulation Modeling ................................................ 37
Outline of Test Program ..................................................................... 38
Regulatory and Water Rights Issues .................................................. 38
Institutional Constraints ..................................................................... 39
Economic Considerations .................................................................. 40
Final Report ....................................................................................... 41
2.3 Phase 2: Field Test Program ............................................................. .41
Baseline Testing ................................................................................. 43
ASR Cycle Testing ............................................................................ 46
Data Collection .................................................................................. 52
Sampling Frequency .......................................................................... 55
ASR Test Program Duration .............................................................. 55
2.4 Phase 3: ASR Wellfield Expansion ................................................... 56
Flow Rate Balancing .......................................................................... 56
Well Spacing and Arrangement ......................................................... 56
Stacking ............................................................................................. 57
Wellfield Layout ................................................................................ 53
2.5 Operations and Maintenance .............................................................. 59
Periodic Change in Operating Mode .................................................. 60
Backflushing to Waste During Recharge ........................................... 60
Disinfectant Residual ......................................................................... 60
Monitoring Data and Reports ............................................................. 61
2.6 Water Supply Planning with ASR ..................................................... 62
Chapter 3
Design of ASR Systems .................................................................................... 63
3.1 Wells .................................................................................................. 65
Casing Materials of Construction ...................................................... 65
Casing Diameter ................................................................................. 69
Cementing .......................................................................................... 70
Selection of ASR Storage Intervals ................................................... 70
Screen Design .................................................................................... 73
Pump Setting ...................................................................................... 73
3.2 Wellhead Facilities ............................................................................ 75
Materials of Construction .................................................................. 75
Pipeline Flushing and Waste Flow Discharge ................................... 76
Trickle Flows ..................................................................................... 76
Sampling Taps ................................................................................... 77
Disinfection of Recovered Flows ....................................................... 78
Cascading Control .............................................................................. 79
Air and Vacuum Relief ...................................................................... 86
Pressure and Water Level Measurement.. .......................................... 87
Flow Measurement ............................................................................ 89
Disinfection and pH Adjustment ....................................................... 91
Pump Considerations ......................................................................... 92
Other ASR Well Site Considerations ................................................. 95
Energy Recovery ................................................................................ 96
3.3 Wellfields ........................................................................................... 97
Dispersive Mixing .............................................................................. 97
Advective Mixing .............................................................................. 99
Chapter 4
Selected ASR Technical Issues ..................................................................... 101
4.1 Recovery Efficiency ........................................................................ 102
Definition ......................................................................................... 102
Water Quality Improvement with Successive Cycles ...................... 104
Water Quality During the Initial ASR Cycle ................................... 108
4.2 Well Plugging and Redevelopment ................................................. 111
Plugging Processes .......................................................................... 112
Measurement Methods for ASR Well Plugging .............................. 118
Normalization of Plugging Rates ..................................................... 120
Source Water Characterization ........................................................ 121
Well Plugging Relationships ............................................................ 124
Plugging Rate Site Investigations .................................................... 124
Redevelopment ................................................................................. l33
4.3 Wellhead Filtration .......................................................................... 136
4.4 Flow Control .................................................................................... 140
4.5 Disinfection Byproduct Reduction .................................................. 142
4.6 Pre- and Post-Treatment.. ................................................................. I 50
Disinfection ...................................................................................... 151
Iron ................................................................................................... 151
Manganese ....................................................................................... 156
Arsenic ............................................................................................. 157
Radon ............................................................................................... 158
Ion Exchange .................................................................................... 159
Hydrogen Sulfide ............................................................................. 160
4.7 Simulation Modelling ...................................................................... 161
ASR Water Supply System Model .................................................. 161
ASR Wellfield Operations Model .................................................... 163
Solute Transport Models .................................................................. 165
Chapter 5
Geochemistry .................................................................................................. 169
5.1 Introduction ...................................................................................... 169
5.2 Water Chemistry .............................................................................. 171
Parameters ........................................................................................ 171
Mass Balance ................................................................................... 177
Water Chemistry Diagrams .............................................................. 178
Hydrogeochemical Processes ........................................................... 180
Changes in Groundwater Chemistry with Depth ............................. 183
Eh-pH Diagrams .............................................................................. 186
5.3 Aquifer Characteristics .................................................................... 191
Physical Characteristics ................................................................... 193
Colored Pictures of Cores ................................................................ 193
Mineralogy ....................................................................................... 194
Cation and Base Exchange Capacity ............................................... 194
Scanning Electron Microscopy (SEM) ............................................ 194
Thin Section Petrography ................................................................ 196
5.4 Geochemical Processes .................................................................... 199
Suspended Solids Clogging ............................................................. 199
Biofouling ........................................................................................ 200
Adsorption ........................................................................................ 203
Ion Exchange .................................................................................... 203
Oxidation .......................................................................................... 204
Dissolution ....................................................................................... 204
Kinetics ............................................................................................ 208
5.5 Geochemical Models ....................................................................... 208
WATEQ ........................................................................................... 209
PHREEQE ........................................................................................ 210
MINTEQ .......................................................................................... 210
EQ3NR/EQ6 .................................................................................... 210
5.6 LaboratoryTesting ........................................................................... 211
Column Testing ................................................................................ 211
Batch Testing ................................................................................... 215
5.7 Field Investigations .......................................................................... 216
Chapter 6
Selected ASR Non-Technical Issues ............................................................. 217
6.1 Economics ........................................................................................ 217
6.2 Water Rate Impacts .......................................................................... 221
6.3 Legal and Regulatory Issues ............................................................ 225
EPA Surface Water Treatment Rule ................................................ 225
EPA Disinfection Byproducts Rule ................................................. 226
EPA Groundwater Rule ................................................................... 226
EPA Underground Injection Control Program ................................ 227
Ownership of the Stored Water ....................................................... 231
Non-Degradation of Groundwater Quality ...................................... 232
Seasonal vs. Long-Term Storage ..................................................... 233
Recovery Percentage ........................................................................ 234
Water Level Impacts ........................................................................ 234
Location for Recovery of Stored Water ........................................... 235
Permit Timing Relative to ASR Feasibility Investigations .............. 235
ASR Education ................................................................................. 236
6.4 Environmental Impacts .................................................................... 236
6.5 Public Involvement .......................................................................... 237
Chapter 7
Alternative ASR Applications ...................................................................... 241
7.1 Introduction ...................................................................................... 241
7.2 Surface Water Storage ..................................................................... 242
Technical Considerations ................................................................. 242
Regulatory Considerations ............................................................... 245
Economics ........................................................................................ 246
Agricultural Applications of Surface Water ASR ........................... 247
7.3 Untreated Groundwater Storage ...................................................... 248
7.4 Reclaimed Water Storage ................................................................ 249
Reclaimed Water Injection Experience ........................................... 250
California Draft Regulations ............................................................ 251
Suggested Regulatory Strategy ........................................................ 253
Chapter 8
Future Directions ........................................................................................... 257
8.1 Technical Developments .................................................................. 258
8.2 Regulatory Issues ............................................................................. 260
8.3 Global Applications of ASR ............................................................ 261
Driving Forces ................................................................................. 261
Constraints ....................................................................................... 262
Opportunities .................................................................................... 262
Chapter 9
Selected Case Studies ..................................................................................... 265
9.1 Peace River, Florida ......................................................................... 265
9.2 Cocoa, Florida .................................................................................. 270
9.3 Marathon, Florida ............................................................................ 275
9.4 Port Malabar, Florida ....................................................................... 282
9.5 Boynton Beach, Florida ................................................................... 286
9.6 Okeechobee, Florida ........................................................................ 289
9. 7 Chesapeake, Virginia ....................................................................... 293
9.8 Swimming River, New Jersey .......................................................... 298
9.9 Wildwood, New Jersey .................................................................... 304
9.10 Kerrville, Texas ................................................................................ 305
9.11 Highlands Ranch, Colorado .............................................................. 309
9.12 Las Vegas Valley Water District, Nevada ....................................... 315
9.13 Calleguas Municipal Water District, California .............................. 318
9.14 Goleta, California ............................................................................. 323
9.15 Pasadena, California ........................................................................ 326
9.16 Seattle, Washington ......................................................................... 328
9.17 Kuwait .............................................................................................. 330
1.1 Estimated annual world water use, total and by sector, 1900-2000 .......... 3
1.2 Ground water recharge ................................................................................ 6
1.3 Location of ASR systems in the United States, 1993 ............................... 12
9.1 ASR Well S-6 monthly recharge and recovery volumes and TDS
concentrations, Peace River, Florida ....................................................... 267
9.2 Location of water supply facilities, Cocoa, Florida ................................ 269
9.3 Water treatment plant and ASR wells, Cocoa, Florida ........................... 270
9.4 Recharge and recovery volumes (8/92-7/93), Cocoa, Florida ................ 273
9.5 Marathon, Florida, ASR location ............................................................ 274
9.6 Observation well chloride concentration vs. time, Marathon,
Florida ..................................................................................................... 278
9. 7 Construction details of the ASR facility, Port Malabar, Florida ............. 282
9.8 Chloride concentrations observed during ASR cycle testing,
Port Malabar, Florida .............................................................................. 283
9.9 ASR well construction and lithology, Okeechobee, Florida ................... 289
9.10 ASR well construction, Chesapeake, Virginia ........................................ 292
9.11 Chloride concentrations during cycles 1 & 2, Chesapeake,
Virginia ................................................................................................... 295
9.12 Core log of Upper PRM aquifer, Swimming River, New Jersey ............ 298
9.13 Construction of ASR Well SR-1, Swimming River, New Jersey ........... 299
9.14 Hydrograph for Arapahoe Well A-6, Centennial Water and
Sanitation District, Highlands Ranch, Colorado ..................................... 308
9.15 Downhole Flow Control Valve ............................................................... 312
9.16 Construction of ASR Wells 16 and 17, Las Vegas Valley Water
District, Nevada ...................................................................................... 314
9.17 Hydrogeologic cross-section, Calleguas Municipal Water District,
California ................................................................................................ 318
9.18 Lithology and well construction, Calleguas MunicipalWater District,
California ................................................................................................ 320
CHAPTER
Introduetion
Slowly they wind athwart the wild,
And while young Day his anthem swells,
Sad falls upon my yearning ear
The tinkling of the Camel-hells.
The Kasidah of Haji Abdu El-Yezdi
Sir Richard Burton
We are many travelers in a land that has limited water resources, and the
distance and direction to the next oasis are not well known. However, the
camel knows the way and will guide us. By storing water when and where
it is available, the camel is an appropriate symbol for a world in which the
incessant increase in demand for water is challenging our ability to meet
this demand.
Water resqurces must be managed more efficiently and wisely if we are
to sustain the needs of a growing world population. The signs are all
around us, for those willing to see:
1
2 GROUNDWATER RECHARGE AND WELLS
• Groundwater supplies 97% of the water used in Saudi Arabia, yet water
levels are declining at such a rate that groundwater reserves will soon be
exhausted, probably within 20 to 50 years.
• About two-thirds of India is underlain by basalt aquifers that supply water
to hand dug wells for domestic and agricultural use. Increasing withdraw-
als are causing these wells to dry up in many areas, creating the need for
deep well pumping equipment and accelerating the rate of water level
decline.
• Groundwater withdrawals in the Denver Basin of Colorado in the U.S. have
caused water levels to decline from near land surface to a depth of almost
275 m (900 ft).
• Wells supplying growing populations in many coastal areas have been lost
to saltwater intrusion directly attributable to increasing groundwater with-
drawals.
• Loss of wetlands and natural areas to urban development and agricultural
production is accelerating the rates of storm runoff and erosion in many
countries, creating pronounced increases in the intensity of flooding and
droughts in downstream areas, and reducing natural recharge to aquifers.
• In Beijing, China, water levels have been dropping about I to 2 m per year
and about one-third of the wells have reportedly gone dry [ 1].
• With a 1992 population of 56 million, Egypt has a renewable water supply
av,eraging only about 82 LPC/day (22 GPC/day). The population of Egypt
is doubling about every 28 years [2].
At the time of Christ, world population has been estimated to have been
approximately 300 million people [3]. It required 1700 years for the
population to double to slightly over 600 million people. By about 1860,
a span of 160 years, it had doubled again to 1.2 billion. The population
reached 2.4 billion by about 1945, doubling within 85 years. It had doubled
again to 4.8 billion by about 1984, 39 years, and is expected to double
again by the end of the century, another 16 years. Annual population
growth rates for the period 1980 to 1988 averaged about 1.7% worldwide,
within a range of 0.3% for Europe to 2.8% for Africa.
Per capita water demands have tended to rise, associated with standards
of living that have improved in many parts of the world. However, per
capita water supplies have fallen rapidly, associated with increasing popu-
lation growth. The widening difference between per capita demand and
supply represents a growing potential for problems and a growing chal-
lenge for water managers. Figure 1.1 shows the estimated annual world
water use between 1900 and 2000. During this period, water use has
increased almost ten times to over 5000 km3/year [4].
Global water supplies are generally believed to be constant. About
40,000 km 3/year constitutes the world's renewable freshwater supply [5].
INTRODUCTION 3
a~~~~~~~~~~~~~~~~~~~~~~~
Figure 1.1 Estimated annual world water use, total and by sector, 1900-2000.
From Shiklomanov, I. A., Global Water Resources, Nature and Re-
sources, Vol. 26, No. 3, 1990. With permission.
I
Basin Recharge Well Recharge
I
Single Purpose Dual Purpose
(Injection Wells) (Aquifer Storage
Recovery - ASR Wells)
Figure 1.2 Groundwater recharge.
The old saying that "there is nothing new under the sun" is also true of
ASR. The Kara Kum Plain is a 311,000 km2 (120,000 miles 2) desert
located on the southeast shore of the Caspian Sea in Turkmenistan, and is
characterized by black sand, alluvial plains interspersed by sand dunes.
Rainfall is usually less than 100 mm (4 inches). During the past several
hundred years, the nomads of this area found that recharge during infre-
quent rainfall in the area occurred only beneath the sand dunes, while a
shallow clay-silt layer and transpiration from desert vegetation prevented
recharge in other areas. The groundwater occurs under water table condi-
tions but is quite brackish except under the sand dunes. To ensure water
supplies, the nomads dug long trenches, extending radially from the sand
dunes and with lengths of up to 2 to 3 km/km2 of tributary area. They were
graded to convey intercepted surface runoff to a central pit excavated in
the dunes. A system of hand-dug wells, cased with locally available brush
wood woven with grass and camel wool, was then constructed surrounding
the central pit, to augment recharge during rainfall events and also to
supply water. Outer wells were utilized to provide poorer quality water for
livestock, while central wells were utilized to meet potable water needs.
With depletion of the stored water between rainfall events, outer wells
would be abandoned as they became too salty, and central wells would be
partially filled in to reduce upwelling of salty water from below, thereby
skimming the residual freshwater in the stored water bubble. Some of
these underground reservoirs were utilized routinely while others were
reserved for use during severe droughts. Because of the highly turbid
recharge water, the wells required cleaning every year [6].
It is perhaps ironic that the term "ASR," first utilized by the author in
1983 to describe the process of underground storage of water using dual-
purpose wells, is also a word that means "capture" in Arabic. Certainly, for
the nomads of the Kara Kum Plain, capturing the limited available rainfall
was absolutely essential for their survival.
In western India, a tribal community has been practicing artificial
recharge for several centuries to obtain drinking water for its members and
livestock. The primary occupation of the tribal people is cattle breeding
and selling milk products. The area is known as Banni and is located in the
northern part of the Kutch District of Gujarat State in western India. The
area is a raised tidal flat, about 5 m above sea level, and covers about 700
km2• Annual rainfall is only 15-20 em, most of which occurs during the
monsoon season from July to September. During the rainy season, runoff
10 GROUNDWATER RECHARGE AND WELLS
United States
Many well recharge studies and field investigations have been con-
ducted during the past few decades; however, few projects are currently
operational. Table 1.1 includes a list of 24 known recharge well projects,
all of which are currently operational. Of these, 20 are ASR projects, all
of which have become operational since 1968. Figure 1.3 shows the
location of ASR facilities in the U.S. as of 1994, including systems in
operation and others in various stages of development.
INTRODUCTION 11
Canada
Year
Operation
location Began Storage Zone Application
• In Operation
o In Development
Israel
Since about 1956, artificial recharge through wells has been an impor-
tant element of the National Water System for Israel. Most of the long-
term operating experience has been in the sandstone aquifer of the coastal
plain and the limestone-dolomite aquifer of central Israel, although some
investigations have also been conducted in the basalt aquifer of lower
Galilee. Recharge has occurred primarily through dual-purpose injection/
production wells (ASR wells), although single purpose recharge wells,
recharge basins, and abandoned quarries are also utilized. The water
INTRODUCTION 13
England
The Netherlands
Australia
21. Diurnal Storage. ASR wells have been used in some areas to store water
at night for recovery during the day, in situations where daytime demands
exceed supply capacity.
22. Fish Hatchery Temperature Control. Seasonal variations in source water
temperature can be used to advantage by recovering and blending ASR
water to meet temperature control requirements.
ASR
Program
Development
The winds blow forth; to earth the quivering lightnings fall,
The plants shoot up; with moisture streams the realm of light.
For all the world ahundallt nourishment is hom,
When by Pwjanya Earth is fertilized with seed.
The rain of heaven bestow, 0 Manas, on us,
Of your strong steed pour forth the streams abundant.
With this thy thundering roar do thou come hither,
And shed the waters as our heavenly father.
With roar and thunder now the germ deposit,
Fly round us with thy water-hearing chariot.
Turn well thy water-skin unloosed downward,
Make, with the waters, heights and hollows level.
Draw the great bucket up and pour it downward,
And let the liberated streams flow forward.
On all sides drench both heaven and earth with fatness;
Let there he for the cows fair pools for drinking.
from the Rig Veda, Book V, 83, Parjanya (c. 1300 BC)
2.1 INTRODUCTION
In some cases, the level of risk is higher than usual, justifying a greater
number of phases along the path to implementation. An increased number
of phases may also be necessary as a result of funding constraints, particu-
larly during field investigations. Certain regulatory, legal, or water rights
issues may be sufficiently important that a separate phase may be devoted
to full investigation of these issues before remaining portions of the
program can continue. In other cases, the level of risk is known to be quite
low as a result of nearby successful ASR experience, justifying moving
immediately into field investigations following an initial conceptual de-
sign effort.
The natural tendency to forego initial investigations and move immedi-
ately into field testing at a selected site tends to be risky. Some ASR
projects have developed in this fashion and have been successful. Others
have encountered significant problems. The problems usually lead to loss
of confidence in the concept of well recharge, as a result of which the
project is halted. The resulting loss of momentum can be hard to over-
come. More careful attention to initial planning details can identify and
resolve many of these issues in advance, thereby minimizing the risk of
failure. Where the penalty associated with failure is low, the higher risk
may be worth taking. However, where the penalty of failure is high,
possibly leading to the need for development of a far more expensive water
supply alternative, it is usually wiser to invest more in the proper initial
development of the ASR program to maximize the chance of success.
The probability of successfully implementing an ASR program can be
enhanced by assembling a multi-disciplinary technical team that includes
a balance of engineers and hydrogeologists with capabilities in the areas
ASR PROGRAM DEVELOPMENT 25
Recharge Objectives
Other ASR objectives are undoubtedly possible. This list may be used
to assess potential primary and secondary objectives. For example, a
community may have a primary objective of storing potable water to meet
seasonal peak demands and thereby defer the need for expansion of water
treatment facilities. Location of the associated aquifer storage recovery
wells could be at the water treatment plant or at any suitable location in the
ASR PROGRAM DEVELOPMENT 27
Water Supply
7,000 25
6,000
f- 20
5,000
---Maximum
f- 15
:aE
.s
&4.000 ()
g
~ 3,000
u:: /Minimum ~verage - 10~u::
I
2,000
I . ..
___ , , ,
-5
.
1,000
Note:
Monthly minimum flow in Peace river. below which no
diversions may occur. is 130 cts (84 mgd. 0.3 mcmd)
except in April and May when it is 100 cts.
Where multiple sources are available, the results can then be compared
between sources to assess their relative suitability for ASR purposes.
20
f'"
25.---------------------------------------------------,
(Based on average monthly flows)
>'o:"_oo loloko Cop~'';"" '""":'
'0
sc 15
01
0
·~
~
010
:2
"E
.!!
0
... 5 Longest period of 22 mgd =50 months
Longest period of 0 mgd = 7 months
6% of the time no withdrawal
35% of the time less than 22 mgd
59% of the time 22 mgd
0 ~L_L~
1 UL-~
1 _j_~
1 ~=r===G====~~~====~~~--~L_~~_L~
1932 1937 1942 1947 1952 1957 1962 1967 1972 1977 1982 1987
which can affect recharge activities. Months when high flows are available
for recharge can also be months when significant water quality issues are
prevalent that would create water treatment difficulties or cause ASR wells
to plug. A thorough scan of recharge water quality records for each source
is required in order to properly assess the potential for problems. An initial
scan on at least one sample obtained during typical recharge months and
flows can be quite helpful in guiding the future direction of the project.
Chapter 9 includes several examples of water quality data for specific ASR
sites, showing the parameters deemed to be of significance at these sites.
Comparison of recharge water quality constituent concentrations with
applicable water quality standards is an important element of the prelimi-
nary feasibility assessment. Where treated drinking water is to be stored,
it is usually sufficient to show that primary and secondary standards are
being met during recharge months. Where other water sources are under
consideration for ASR storage, such as untreated or partially treated sur-
face water, untreated groundwater, or reclaimed water, then comparison
with applicable water quality standards can provide a basis for regulatory
evaluation of the proposed ASR project, as discussed further in Chapter 7,
Alternative ASR Applications. Appendix A includes 1993 water quality
standards for the U.S. Environmental Protection Agency, the European
Community, and the World Health Organization.
For the Peace River water supply example discussed above, potential
diversions from the river were further constrained by the algal content of
the river water. Analysis revealed that there are several months during the
year when water may be available for diversion, considering only quantity
criteria; however, the quality is so poor that it cannot be readily treated.
The probability of not diverting due to algal content varied from zero in
September through December to one-third in March. Furthermore, expe-
rience at that site suggests that this is not easily predicted, occurring
somewhat randomly during certain months of the year and averaging about
17% of the potential diversions for the entire year. This analysis had a
pronounced effect upon the initial conceptual design of the ASR system
and the other facilities at the Peace River site. Subsequently, facilities were
constructed and additional facilities are planned so that this water can be
diverted and treated rather than lost.
An important water quality consideration is the suspended solids con-
tent of the recharge water source. Invariably this data is not available for
Phase 1 investigations. As discussed in Chapter 4, well plugging and
redevelopment is an important ASR technical issue, the resolution of
which includes field data collection to gather information regarding total
suspended solids content in the recharge water, how these solids vary with
time and with flow, and what materials contribute to these solids. In some
30 GROUNDWATER RECHARGE AND WELLS
Water Demand
Municipal water systems are usually designed to meet peak day de-
mands during some future year. The typical ratio of peak day to average
annual demand is about 1.3 to 2.0, although ratios as high as 5.0 are
known. Consequently, it is not uncommon for water systems to have a
substantial amount of idle capacity during much of the year. This capacity
can be utilized for treatment and storage of water during off-peak months,
using ASR and other recharge facilities.
Figure 2.3 shows the record and projection of average and maximum
day water demand for Evesham, NJ, showing the effect of adding one ASR
well on extending the useful capacity of existing water supply and treat-
ment facilities. Typically, an ASR system enables a water utility to meet
maximum day demands with water supply and treatment facilities sized to
10~----------------------------------------------------.
1.7 mgd
ASR Expansion•
1.7 mgd
ASR Expansion*
Average Day
Demand
• Includes 1.3 mgd from PRM and 0.4 mgd from Mount Laurel Aquifers
2L-~--~--~--~--~~L-~--~--~--~--~--~--~~~
Figure 2.3 Supply-demand relationship with new ASR facilities, Evesham Mu-
nicipal Utilities Authority, New Jersey.
32 GROUNDWATER RECHARGE AND WELLS
Hydrogeology
The cost of conducting this work can be weighed against the perceived
risk and cost of potentially plugging and losing use of a recharge facility.
Where the risk is low, or the value of the facilities at risk is not too great,
then this portion of the assessment can probably be deferred or minimized.
Otherwise, coring and geochemical analysis should be conducted during
34 GROUNDWATER RECHARGE AND WELLS
Processes
Site Selection
For most ASR projects, the site for the Phase 2 test facilities is best
located either at or close to the water treatment plant, or at some point in
the distribution system where major facilities are already in place, such as
a pumping station or a ground storage reservoir. This provides increased
likelihood that qualified personnel will be available during the test pro-
ASR PROGRAM DEVELOPMENT 35
Conceptual Design
Assuming that ASR is selected as the recharge process and one or more
alternative sites have been identified, advantages and disadvantages of
each site can be considered, and criteria can be developed and applied for
recharge site selection. Conceptual design of facilities to achieve recharge
objectives can then be conducted. Usually it is sufficient to develop a
preliminary conceptual design for Phase 2 test facilities, with some con-
sideration for the probable layout of the ultimate ASR site to ensure that
36 GROUNDWATER RECHARGE AND WELLS
A is the decision
whether to utilize an well
purposes or to construct new facilities designed
but at increased cost for the test program. -'-'"''""'"F.
and abandoned are frequently and
ered for test program purposes, for reasons.
not seldom constructed
may not be in a condition suitable for purposes.
As discussed subsequently in Chapter 3, ASR wells have a
design that is different from either production or injection wells. Where
zone water quality is fresh and potential geochemical problems are
production well may be very similar to ASR well
However, when the storage zone water quality differs from water
and where geochemical issues are of concern, the ASR well
will tend to be different from the well
selection between using an existing or a new
some care if optimum well performance is to be achieved.
Where an existing well is it will be necessary to
the condition of the well to testing. Whether this is done Phase
1 or at the of Phase 2 is a site-specific decision.
exercise may include video camera logging of the
activities as wire brushing the screen and casing while pumping the
acid treatment; disinfection; and a second video log to confirm results. In
old wells that are equipped with oil-lubricated pumps, it is common to find
a layer of oil floating in the casing, which should be removed prior to ASR
testing operations.
It is important that program objectives not be compromised by a
sion to utilize existing facilities that are unsuitable for the purpose. Once
an ASR test well becomes severely plugged or has other major problems
during testing, the test program tends to lose support. It is then very
difficult to regain support by pointing out that further investment in new
ASR PROGRAM DEVELOPMENT 31
test facilities would have been more likely to achieve success. Wells are
usually abandoned or not utilized for a reason: due to construe-
age, sand production, or other water quality problems that can se-
""''"J'A the conduct of successful recharge testing. Where
for great care has to be exercised to
if possible, correct shortcomings m design or construction
affect success of the program.
These issues are different in every part of the U.S. Each state has its own
requirements and procedures. The federal government (U.S. Bureau of
Reclamation, U.S. Geological Survey) also is involved in recharge activi-
ties through funding of local demonstration programs and state water
ASR PROGRAM DEVELOPMENT 39
Institutional Constraints
• established policies that were developed many years ago without consider-
ation of ASR potential opportunities, and
• numerous other institutional constraints
Failure to carefully evaluate and address these often subtle issues can
easily lead to recharge program failure or extended delay. Conversely, use
of the Phase 1 feasibility report as a tool to elicit political and institutional
support for the recharge program can lead to constructive input and en-
hanced likelihood of ultimate success.
In some cases institutional or regulatory issues will create the need for
public involvement in the planning and implementation of the recharge
program. This can occur in several ways but usually involves meetings
with advisory committees or at regulatory hearings during development
and permitting of the recharge program. The Phase 1 report can be a
valuable tool at such meetings to help present the overall program in an
unambiguous way and thereby dispel misconceptions that frequently arise.
Economic Considerations
Final Report
Once the decision is made to move ahead with a Phase 2 test program,
the first step usually is to design and construct the ASR test facilities.
These should be designed at full scale, capable of recharging and recov-
ering at whatever rates are reasonably available from an efficient well
penetrating the selected aquifer. Testing in a small diameter test well
necessitates recovery rates that are lower than those that would ultimately
be expected from a full size well. Conclusions regarding potential seasonal
42 GROUNDWATER RECHARGE AND WELLS
volumes and associated unit costs therefore tend to be biased on the high
side. In some situations, such as storage of freshwater in brackish aquifers,
the extrapolation of results from small-sized facilities can lead to incorrect
conclusions regarding ASR recovery efficiency and cost-effectiveness
compared to results from properly-sized test facilities. Design issues will
be addressed in greater detail in Chapter 3, Design of ASR Systems.
Construction issues are similar to other well construction projects, with
the possible exception of the greater amount of hydrogeologic data col-
lected during construction. Sometimes this includes collection of cores for
detailed laboratory analysis to determine mineralogy, geochemistry, and
associated hydraulic properties, as discussed in Chapter 4, Selected ASR
Technical Issues, and Chapter 5, Geochemistry. Chapter 6 discusses permit-
ting issues associated with construction and operation of ASR facilities.
Interval pumping tests are sometimes conducted at several different
depths during construction of open hole wells, in order to estimate the
productive intervals of the aquifer for correlation with geophysical
and drill cutting descriptions. Geophysical logging is conducted after pilot
hole drilling in order to establish casing and screen intervals. For screened
wells, caliper logs may be obtained after the pilot hole is reamed and prior
to setting the screen, while flowmeter logs are obtained after the screen is
installed and developed. For open hole wells, flowmeter and caliper logs
are obtained following well completion. The baseline distribution of pro-
duction with hole depth or screen interval is useful as a reference point
against which to compare subsequent logs and thereby estimate changes in
the flow distribution due to ASR operations. Upon completion of ASR
facilities construction, the test program begins.
Design of the test program reflects a careful assessment of the many
issues of concern at each new site. Well plugging is always a primary
issue; however, others of importance may include the following:
Baseline Testing
The first part of the test program includes baseline hydraulic and water
quality testing prior to initiating significant recharge activities. This pro-
vides a reference point against which future results may be compared.
Baseline hydraulic testing usually starts with a standard step drawdown
pumping test in order to establish well and formation loss coefficients and
well efficiency. Following water level recovery, a long duration pumping
test should be conducted in order to estimate aquifer hydraulic character-
istics in the vicinity of the ASR well. Duration of the test depends upon
local experience in obtaining reliable estimates of aquifer hydraulic char-
acteristics. Typical durations are about one day, sometimes longer. If
observation wells are present, they should be incorporated in the pumping
test to better define aquifer transmissivity and estimate storativity. Obser-
vation well data are frequently more useful since it is less affected by
variations in the pumping rate that affect drawdowns in the pumping well.
The test establishes aquifer hydraulics in the vicinity of the well before
recharge commences.
Upon completion of the long-term pumping test and associated recov-
ery of water levels to background, a step-injection test is usually conducted
to characterize water level response in the ASR well under reverse condi-
tions from the previous step-drawdown test. Recharge during this step-
injection test occurs at three different increasing rates, generally bracket-
ing the expected recharge rate for the well. Each recharge rate step is of
the same short duration, such as 2 to 4 hours. Water level response to this
test characterizes the baseline water level response of the well in the
presumed absence of significant plugging. At any subsequent point in
time, this test can be repeated to evaluate whether plugging has occurred
between the two tests. This is discussed further in Chapter 4, Selected ASR
Technical Issues.
An important element of the baseline step-injection test is that the
recharge water should be allowed to flow to waste near the wellhead at the
planned recharge rate for a few minutes prior to recharge down the well or
until any solids in the recharge water have been flushed from the system.
Depending upon the length and the normal flow rate in the tributary piping,
44 GROUNDWATER RECHARGE AND WELLS
flow reversal in the piping during ASR recharge can sweep a substantial
volume of solid material down the well, causing immediate onset of
plugging if these solids have not first been purged from the system.
Baseline water quality characteristics are also determined during this
initial part of the test program. Samples are collected usually at the
beginning, middle, and end of the long-term pumping test to fully charac-
terize water quality in the ASR well prior to recharge. The last sample
usually receives a complete water quality analysis, including EPA primary
and secondary drinking water standards and several other parameters, as
suggested in Table 2.2. The initial and middle samples collected during
this long-term pumping test are analyzed for a smaller range of parameters
in order to estimate whether any trend in water quality occurs during the
test If a trend is apparent from field data such as chloride, conductivity,
or pH, it may be appropriate to extend the pumping test until such time as
equilibrium water quality is apparent.
Sometimes the situation arises where the storage zone is separated
a thin or poorly defined confining layer from an overlying or underlying
highly transmissive aquifer containing water of very poor quality. The
adequacy of this confining layer may be of some concern. In particular, if
the time required for poor quality water to move through the confining
layer under the head differential imposed during seasonal recovery is
shorter than the expected ASR operational recovery period, then deterio-
rating recovery water quality will tend to define the upper limit of ASR
recovery duration and volume, regardless of the volume stored.
Figure 2.4 shows the ASR initial test wells and storage zone at Cocoa,
FL. Chloride concentration in the storage zone is about 400 mg/L; how-
ever, an underlying aquifer, not penetrated by the ASR well, has a chloride
concentration of 1320 mg/L. The confining layer separating the aquifers
was approximately 36ft thick and was of uncertain integrity. Following
initial pump testing and cycle testing to determine storage zone hydraulic
characteristics and water quality response to ASR operations, it was evi-
dent that upflow was occurring from the underlying aquifer through the
confining layer during recovery. The ASR well and production zone
observation well were then plugged back to a shallower depth, with
cement, to reestablish the integrity of the lower confining layer. The
shorter open hole interval of the ASR well was acidized to restore the
production capacity lost when the well was plugged back, following which
a 90-day pumping test was performed at the design recovery rate for the
ASR well. Chloride concentration during the test remained steady, sug-
gesting that future ASR operations with recovery periods of up to 90 days
would not be likely to experience reduction in water quality due to up-
welling of brackish water from below.
)>
(f)
Depth in
PZ-l FH SMW-1 DMW-1 JJ
Feet
"U
JJ
o~~--~----~m 0
G)
JJ
)>
s:
0
100 m
<
m
r
0
"U
s:
m
200 z
-1
24-lnch
Hole
300 Chloride= 380-420 mg/L
8-lnch
Casing~
400
500
516Feet-
8-lnch
600 Chloride= 1300-1400 mg/L
Open Hole
to 595 Feet
Figure 2.4 ASR initial test facilities and storage zone, Cocoa, Florida.
O'n
""'
46 GROUNDWATER RECHARGE AND WELLS
Another ASR test site at Tampa, FL, showed inadequate lower confine-
ment from a similar test, as evidenced by breakthrough of poor quality
water from an underlying highly transmissive aquifer with a total dis-
solved solids (TDS) concentration of 8000 mg/L. The storage zone had
been acidized to improve its low initial yield, resulting in a dramatic
increase in specific capacity. Unfortunately, the acid opened up a channel
through the lower confining layer, which was about 21 m (80ft) thick. The
breakthrough of poor quality water during recovery occurred after a given
volume of recovery, essentially regardless of the volume recharged. In this
case, plugging the bottom of the well with cement may possibly help to
restore adequacy of the lower confining layer; however, this may also
substantially reduce well yield. Relocation of the ASR well would then be
necessary.
ASR system. A brief interval between the first and second cycle is some-
time useful to obtain and review laboratory water quality results and
hydraulic data, permitting adjustment of the testing plan for subsequent
cycles or adjustment of the ASR facilities, if appropriate.
If water quality difference between stored and native water is small and
there are no significant concerns regarding geochemical reactions, then a
small number of long cycles is appropriate in order to focus upon plugging
rates and backflushing frequency required to maintain recharge rates. A
minimum of three cycles is usually appropriate in such situations, with the
third approximating an operational recharge duration. Recovery of 100%
of the stored water volume in each cycle after the first is a reasonable
target, although greater or lesser recovery volume may be appropriate in
some situations.
If there is a significant water quality difference between stored and
native water, a larger number of cycles will be required. After the first
cycle, the next three cycles have the same recharge volume and storage
period in order to demonstrate improvement in recovery efficiency with
successive identical cycles. Subsequent groups of three or more cycles
tend to have larger volumes and, in some cases, may incorporate storage
periods as discussed below. The total number of cycles may be in the range
of 4 to 10. Recovery occurs to a target water quality concentration in each
cycle. The water not recovered in each cycle forms a buffer zone to
improve quality during the subsequent cycle. The target constituent con-
centration may or may not be a potable water quality standard, depending
upon whether any blending will occur between the recovered water and
water from other sources prior to consumption.
If there is a real concern regarding potential geochemical reactions, care
should be taken to avoid shocking the formation with a sudden change in
quality. Furthermore, storage time should be built into the test program
since some reactions such as manganese dissolution require several days
or weeks to occur. Geochemical plugging reactions that occur several feet
away from the well screen have little effect upon well hydraulics, while
those occurring close to the well can have substantial adverse effects.
Hence, in these cases it is appropriate to design the ASR test cycles in such
a way that stored water is never fully recovered, always leaving a small
buffer around the well, which tends to build with successive cycles.
Where potential geochemical reactions are an issue, it is advisable to
run a larger number of small cycles to demonstrate control over geochemi-
cal issues near the well, before moving on to larger operational cycles. It
would not be unusual to run 6 to 10 cycles in situations with complex
geochemical issues.
48 GROUNDWATER RECHARGE AND WELLS
Initial ASR cycles are usually performed with the recovered water
discharged either to waste or to retreatment. Water from subsequent cycles
during the test program can usually be returned to the distribution system
once adequate water quality has been demonstrated and the system is
permitted. Frequently the last test cycle is designed so that the volume to
be recovered is large, and recovery occurs during a time of peak demand,
so that the water meets a local need.
The effect of storage time on system performance is always of interest.
However, it is quite rare for this to be a significant parameter in ASR
operational performance. One situation where the effect of storage time is
significant is where the storage zone TDS concentration is sufficiently
high that density stratification is significant. A difference between re-
charge and native water TDS concentration of about 5000 mg/L or higher
is probably a threshhold above which this should be considered. A second
situation would be where the regional or local hydraulic gradient around
the ASR well is sufficiently steep, and the aquifer transmissivity suffi-
ciently high, that there is real concern that the stored water will rapidly
move away from the well prior to or during recovery, such that poor
quality native water would then be recovered. Finally, storage time can be
significant in situations that are geochemically complex. For example,
manganese production requires several days or weeks to develop, as
discussed above. Other than these situations, it is better to spend the time
more productively, recharging and recovering water and gathering data.
Consideration of these guidelines will reveal that priorities need to be
established at the outset of the test program design, since some of the
guidelines are incompatible with others. For example, it is not advisable to
recover 150 to 200% of the stored water during the first cycle if the initial
data suggests that adverse geochemical reactions may be occurring. For
low permeability storage zones in brackish sand aquifers with no significant
geochemical sensitivity, definition of plugging rates and redevelopment
frequencies may take precedence over demonstration of recovery efficiency
improvement in successive cycles, moving the test program in the direction
of a small number of long cycles instead of a larger number of short cycles.
Tables 2.1, 2.2, and 2.3 present the actual test programs implemented
at three sites, illustrating a range of issues to be resolved. Each site is
discussed in greater detail in Chapter 9, Selected Case Studies.
In the first case, Table 2.2 shows the test program for Marathon, FL.
This ASR site includes a geochemically insensitive sand aquifer contain-
ing water with a TDS concentration of 39,000 mg/L. The storage zone is
confined, and the ASR test well has a low specific capacity of about 3 G/
min/ft. The site provides an emergency water supply for the Florida Keys,
)>
(f)
JJ
'"0
JJ
0
GJ
JJ
)>
;;;;:
0
TABLE 2.1 CYCLE TEST PROGRAM: MARATHON, FLORIDA m
<
m
r
Volume (MG) Duration (Days) (a) 0
'"0
Cycle No. Recharge Recovery "/o Recharge Storage Recovery Total ;;;;:
m
z
-1
1 4.5 5.1 33 17 0 19 32
2 9.7 3.5 30 42 34 1i 87
3 5.3 4.2 68 27 0 16 43
4 3.6 2.8 72 14 0 9 23
5 15.0 6.5 43 51 39 26 116
6 15.1 7.7 51 56 36 43 135
7 15.8 8.9 55 56 35 30 i 21
8 15.4 i 0.1 65 76 21 34 131
9 15.0 10. i 65 54 0 44 98
10 15.3 8.6 56 56 35 31 122
11 14.0 10.4 71 63 81 (b) 25 169
Note: (a) Testing conducted 1990 to 1993. (b) Trickle flow of about 50 G/min during last 57 days of storage
period. Total recharge plus trickle flow volume = 19.2 MG.
J>,
(!)
50 GROUNDWATER RECHARGE AND WELLS
Average
Gallons Recharge/
Recharged/ Recovery
Recovered Rate
Begin End Total Time (millions) (G/min)
Cycle 1
Recharge 4/2/91 4/5/91 3 days, 1 hr. 2.93 669
Storage 4/5/91 4/7/91 2 days
Recovery 4/7/91 4/9/91 2 days, 7 hrs. 2.86 867
Cycle 2
Recharge 4/15/91 5/14/91 29 days, 3 hrs. 24.9 595
Storage 5/14/91 6/13/91 30 days
Recovery 6/13/91 7/3/91 20 days 25.0 868
Volume Volume
Injected Recovered "/o
Cycle Dates (MG) (MG) Recovered
Table 2.4 shows the test program for the ASR facility at Swimming
River, NJ. The storage zone is a confined, clayey sand aquifer that is
geochemically complex and sensitive, including native water iron concen-
trations exceeding 13 mg/L and manganese sources in the storage zone.
The site also has moderate transmissivity of 509 to 658 m2/day (41,000 to
53,000 G/day/ft). Seven ASR test cycles were conducted to initially re-
solve plugging issues and then to address iron and manganese geochemical
issues. As discussed subsequently in Chapter 4, ASR testing at this site
was preceded by a pretreatment phase to condition the aquifer around the
well. Initial resolution of these complex issues was successfully resolved
through small volume cycles in which reactions occurred reasonably close
to the well. It was then possible to move on to larger cycles.
A characteristic of ASR test programs is that utility operational require-
ments often affect the planning or implementation of the testing plan. Water
may not be available for recharge at the rate or time desired. Water may be
required from recovery to meet peak demands at a time that may be inap-
propriate from the limited perspective of the test program. Mechanical and
Frequecy
Parameter Recharge Recovery Daily Weekly
Data Collection
END INJECTION
Figure 2.5A Cycle testing data collection form, Port Malabar, Florida.
Depth I<>
Elapsed Water Recovery Speclllc Volume Alkalinity lli!ardn ..s
Time ll.eoovory (+ =abv. Ml') Rat. Capacity Roooverod ll'orcenl Sample Chloride Cood. "II'" "T" Too.! Ca Turbidity lll2S Sulf&lo
Date Time (hrs) Meter (-=below Ml') (!jpm) (grmln) !pi) Recovery Number (mgn) (mmh/cm) pll plli (mgil) (mgm (mg/1) (mg/1) (n!u) (mg/1) (mg/1)
08/09/88 7 43 16.70 700500 -48.63 700 10.9 700,500 20.15 C3RR2.1 185 676 8.3 8.0 40 190 148 0.22 51
08/09/88 14 20 23.32 975000 -48.15 695 10.9 975,000 28.05 C3RR2.2 186 712 8.4 8.0 40 192 !52 0.30 0.00 52
08/09/88 19 50 28.82 1160000 -%.50 695 11.2 1,160,000 33.37 C3RR2.3 180 7!2 8.5 8.0 40 188 156 0.34 0.00 50
08/10/88 8 20 41.32 1689500 ·50.96 725 10.9 1,689,500 48.60 C3RR3.l 195 730 8.2 7.9 0 44 208 164 0.31 0.00 54
08/10/88 13 10 46.15 !870680 625 1,870,680 53.81 C3RR3.2 200
08/10/88 15 20 48.32 1960000 680 1,960,000 56.38 C3RR3.3 198 734 8.3 7.9 0 44 204 168 0.32 0.00 55
08/1088 19 26 52.42 2123000 -43.50 660 11.1 2,123,000 61.07 C3RR3.4 216 767 8.2 8.0 0 44 220 160 0.38 0.00 60
08/11188 8 15 65.23 2672500 -51.43 730 10.9 2,672,500 76.88 C3RR4.1 250 8% 8.2 7.8 0 60 252 180 1.4 0.00 67
08/11/88 10 48 67.78 2786000 730 2,786,000 80.14 C3RR4.2 2%
08111/88 12 45 69.73 2873245 -49.21 705 !0.9 2,873,245 82.65 C3RR4.3 265
08/11/88 14 41 71.67 2910000 700 2,910,000 83.71 C3RR4.4 265
08111/88 15 45 72.73 2952500 700 2,952,500 84.93 C3RR4.5 275
08/11/88 IS 54 72.88 2961000 2,961,000 85.18 C3RR4.6 947 8.2 7.9 0 44 268 196 0.48 0.00
08111/88 16 45 73.73 2998000 -45.70 725 11.8 2,998,000 86.24 C3RR4.7 263 962 8.2 7.8 0 56 264 184 0.39 0.00 71
08111/88 17 0 73.98 3001985 3,001,985 86.36
END RECOVERY
Figure 2.58
ASR PROGRAM DEVELOPMENT 55
Sampling Frequency
Depending upon the range of issues addressed during the test program,
the duration may extend for as short as about 3 months and as long as 2
years or more. Short programs are appropriate in situations where no
geochemical or water quality issues are involved and where plugging does
not appear to be significant in early testing. On the other hand, where the
56 GROUNDWATER RECHARGE AND WELLS
There are a few factors to consider in the design of an ASR wellfield that
are quite different than for a conventional wellfield. Failure to consider these
factors can reduce the value of an ASR system to meet water demands.
For an ASR system that primarily meets seasonal water supply needs,
the annual volume of water stored will create a bubble of stored water
ASR PROGRAM DEVELOPMENT 57
around the well, with a diameter typically in the range of 100 to 300m (328
to 984 ft). If well spacings are less than this diameter, the storage bubbles
from adjacent ASR wells should tend to coalesce. Over a period of several
cycles, this should tend to improve recovery efficiency, compared to a
wellfield with spacings such that no coalescence occurs. As discussed
subsequently in Section 3.3, Wellfield Design, well arrangements includ-
ing a central well should provide improved recovery efficiency. The
tradeoff between increased annual power costs for recharge and recovery
in a "clustered" ASR wellfield with closely-spaced wells can be compared
with the expected improvement in recovery efficiency, based upon Phase
2 results and modeling, as discussed in Chapter 4. A decision can then be
made regarding appropriate well spacing. In general, closer well spacing
will be appropriate in more brackish aquifer systems, while conventional
well spacing criteria will tend to govern wellfield design in situations
where groundwater quality is fresh.
Stacking
Tampa !l ~
400 Storo e Zone
~
ro
,.~
.A
·a Confining Layer It '~
Suwanee p_
.<::: Storage Zone
a soo ~·
c
~
~()
1,000
Ocala Confining Layer !l
1,200 - !l
1..9.
~~
1,400 -
Avon
Park Storage Zone
Figure 2.6 Stacking ASR storage in multiple zones, Peace River, Florida.
Wellfield Layout
New Pipeline
lOkm
This occurs typically two to four times per year as the system changes
from recharge to recovery and back again. Where ASR wells are used to
meet peak demands in areas where diurnal and weekly variations are more
significant than seasonal variations, more frequent changes in operating
mode will occur. Automated controls facilitate such The pro-
cedure at each well may need to include flushing lines to waste for a few
minutes at each wellhead, depending upon the amount of rust, and
other solid material in the piping. Proper selection of construction mate-
rials can substantially ease this operating requirement.
to
Disinfectant Residual
Monitoring
Monitoring requirements for ASR systems depend upon local needs and
regulatory requirements at each site. It is vitally important to accurately keep
track of flow rates and cumulative volumes at each well and for the entire
ASR wellfield. Regular calibration of flowmeters is therefore essential.
Periodic measurement of water levels in ASR and monitor wells is also
important, in order to check for well plugging rates, pump efficiencies,
effect of ASR operations upon water levels in adjacent wells, and other
62 GROUNDWATER RECHARGE AND WELLS
Design
of
ASR
Systems
Once there was a great drought. The rain stopped falling and the Earth became
dry. Finally the streams themselves stopped flowing. There was a village of
people who lived by the side of a stream, and life nmv became very lzardfor them.
They sent someone upstream to see why the stream had stopped. Before long, the
man came back.
"There is a dam across the stream," he said. "It is holding back all of the
water. There are guards on the dam. They say their chief is keeping all the water
j(n· himself:''
"Go and beg him.fiJr water," said the elders of' the 1·illage. "Tell him we are
dying without water to drink." So the messenger went back again. When he
returned, he held a bark cup filled with mud.
"This is all the Vl'ater their chief will allow us to have," he said.
Now tlze people were angry. They decided to fight. They sent a party of
warriors to destroy the dam. But as soon as the warriors came to the dam, a great
monsta rose out of the water. His mouth vt·as big enough to swallow a moose.
His belly was huge and yellow. He grabbed the warriors and em shed them in his
long .fingers, which were like the roots of cedar trees. Only one warrior escaped
to come back to the people and tell them what happened.
"We cannotfight a monster," the people said. They were not sure what to do.
Then one of the old chief:' spoke. "We must pray to Gitchee Manitou," he said.
''Perhaps he will pity us and send help." Then they burned tobacco and sent their
prayers up to the Creator.
63
64 GROUNDWATER RECHARGE AND WELLS
Their prayers were heard. Gitchee Manitou looked down and saw the people
were in great trouble. He decided to take pity and help them and he called
Koluscap. "Go and help the people," Gitchee Manitou said.
Koluscap then went down to the Earth. He took the shape of a tall warrior,
head and shoulders taller than any of the people. Half of his face was painted
black and half was painted white. A great eagle perched on his right shoulder
and by his side two wolves walked as his dogs, a black wolf and a while wolf. As
soon as the people saw him they welcomed him. They thought surely he was
someone sent by the Creator to help them.
"We cannot afford you anything to drink," they said. "All the water in the
world is kept by the monster and his dam. "
"Where is this monster?" Koluscap said, swinging his war club, which was
made of the root of a birch tree.
"Up the dry stream bed, " they said.
So Koluscap walked up the dry stream bed. As he walked he saw dried up and
dead fish and turtles and other water animals. Soon he came to the dam, which
stretched between two hills.
"I have come for water," he said to the guards on top of the dam.
"Give him none, give him none!" said a big voice from the other side of the
dam. So the guards did not give him water.
Again Koluscap asked and again the big voice answered. Four times he made
his request, and on the fourth request Koluscap was thrown a bark cup half-full
of filthy water.
Then Koluscap grew angry. He stomped his foot and the dam began to crack.
He stomped his foot again and he began to grow taller and taller. Now Koluscap
was taller than the dam, taller even than the monster who sat in the deep water.
Koluscap's club was now bigger than a great pine tree. He struck the dam with
his club and the dam burst open and the water flowed out. Then he reached down
and grabbed the water monster. It tried to fight back, but Koluscap was too
powerful. With one giant hand Koluscap squeezed the water monster and its eyes
bulged out and its back grew bent. He rubbed it with his other hand and it grew
smaller and smaller.
"Now," Koluscap said, "no longer will you keep others from having water.
Now you'll just be a bullfrog. But I will take pity on you and you can live in this
water from now on. " Then Koluscap threw the water monster back into the
stream. To this day, even though he hides from everyone because Koluscap
frightened him so much, you may still hear the bullfrog saying, "give him none,
give him none. "
The water flowed past the village. Some of the people were so happy to see the
water that they jumped into the stream. They dove so deep and stayed in so long
that they became fish and water creatures themselves. They still live in that river
today, sharing the water which no one person can ever own.
Mic Mac and Maliseet Indian story, Nova Scotia [1]
DESIGN OF ASR SYSTEMS 65
3.1 WELLS
ASR wells have certain unique features in their design that differentiate
them from production wells or injection wells. When completed, the wells
may often be similar; however, the design process is different, and the end
results may also be different, depending upon conditions at each ASR site.
In this section, the ASR well design approach is discussed, addressing
those features that differentiate ASR wells from other wells.
ASR wells generate rust from steel casings to a greater extent than either
production or injection wells, due to the increased surface area subject to
wetting and drying during recharge and recovery. This is particularly true
for brackish water storage zones. This rust flows down the well during
recharge, contributing to plugging of the well. Solids present in the re-
charge water are usually more significant causes of ASR well plugging
than is rust. However, for low permeability aquifers, the increase in
plugging potential due to rust can be unacceptable in some cases, particu-
larly where frequent backflushing to waste is perceived as an operating
problem to be avoided if possible. During recovery or backflushing rede-
velopment, the rust combines with other solids carried into the well during
recharge and is pumped from the well, either to waste or into the wellhead
piping system.
For ASR wellfields located at water treatment plants, it is not uncom-
mon for the water pumped from the well at the beginning of recovery to
be conveyed back to the treatment process for retreatment. Duration of this
period may typically range from about 10 min to 2 hours or more. Once
the rust and other particulates have been flushed from the well, the water
can then be diverted directly into the treated water distribution system
following disinfection.
For ASR wells located other than at water treatment plants, the only
option is to waste this water to a nearby drainage system or sewer line. The
pumping rate at such times may be slightly greater than the design recov-
ery rate for the well, since the pump is usually pumping against a lower
head than normal and therefore producing more water. This is good in that
it helps to purge solid material from the well; however, disposal of water
for an extended period at such rates is sometimes a problem. In residential
areas lacking storm drainage networks, ASR backflushing operations can
cause temporary, localized flooding of streets and homeowner opposition
66 GROUNDWATER RECHARGE AND WELLS
PVC Casing
The use of PVC casings at the Cocoa and Marathon, FL, ASR sites has
been effective, in conjunction with appropriate selection of wellhead
piping materials, in keeping the production of solid materials during
recharge and recovery to an acceptable minimum. At each site, water is
clear and meets all applicable drinking standards within about 20 min after
the beginning of recovery.
A factor to consider during installation of PVC casings is that the
density of the PVC is approximately the same as the density of drilling
mud. Consequently, it may be necessary to push the casing down the hole
prior to cementing, particularly if the drilling fluid weighs more than about
12 lbs/G, or has a specific gravity greater than about 1.44.
These have not yet been utilized for ASR wells, primarily due to the
availability of less-expensive alternatives. It is reasonable to expect that
use of these casing materials will occur sooner or later for an ASR system,
in order to meet site-specific needs.
ASR system operating costs are discussed in Chapter 6. At such time as
we have greater understanding of the annual costs associated with periodic
backflushing for well redevelopment, and for initial recovery to waste to
remove solids, the tradeoff between investment in higher cost well casing
and reduced operating costs will become more clear. Until that time, it
seems wise to seek reasonable opportunities to minimize solids produc-
DESIGN OF ASR SYSTEMS 69
Steel Casing
Casing Diameter
For ASR wells, casing diameter should be no different than for normal
production wells, except when one or more injection tubes are used for
70 GROUNDWATER RECHARGE AND WELLS
Cementing
Casings in ASR wells should be cemented from the bottom of the casing
to ground surface to ensure an adequate seal against flow movement
outside the casing through possible channels opened during construction.
At three known ASR sites in existing uncemented water supply wells, the
pressures occurring during recharge caused upward flow around the out-
side of the casing. At one site this created flow at the surface. At another
site the result was formation of a sinkhole adjacent to the well. The third
site experienced downward movement of surficial sands into the underly-
ing limestone production interval, causing a severe solids problem in the
well. Cementing is normally a desirable practice to prevent production
well contamination from adjacent land use activities; however, for ASR
wells there are additional hydraulic reasons that apply due to the cyclic
operation.
The simplest case is one in which the ASR storage zone under consider-
ation contains water of similar quality to that which will be recharged, and
DESIGN OF ASR SYSTEMS 71
has no potential geochemical issues. In such a case, the ASR well design will
tend to be similar to a conventional production well design. If screened, the
screen length will tend to be longer to maximize recharge efficiency and to
minimize the rate of plugging. If open hole, the hole length will tend to fully
penetrate the production interval for the same reason.
In situations where the storage zone is brackish, or contains water of
such quality that mixing is to be minimized, the selection of the storage
interval requires greater care. Thin intervals that have excellent vertical
confinement are best suited for minimizing mixing. Figure 3.1 shows a
geologic cross-section for the Marathon ASR test well, FL, which success-
fully stores treated drinking water for emergency water supply purposes.
Storage is in a confined sand production interval 11 m (40ft) thick and
containing seawater with a total dissolved solids (TDS) concentration of
39,000 mg/L. This site is discussed further in Chapter 9. In less extreme
cases of water quality difference, thicker storage intervals with less con-
finement may be sufficient to provide the desired recovery efficiency.
Where the choice of storage intervals is limited, well and wellfield
design can, to some extent, adapt to the limitations imposed by nature.
Multiple wells can provide ASR development of a zone that has sufficient
storage volume capacity but low yield to individual wells. The cost of
additional wells is frequently small when compared to the cost of alterna-
tive storage approaches.
Where the storage zone has great thickness or poor confinement and
contains poor quality native groundwater, acceptable recovery perfor-
mance may sometimes be achieved by operating at high rates and long
durations during recharge. The volume stored may then be sufficient to
displace the poor-quality water away from the well, both vertically and
laterally, so that a useful recovery volume can be achieved during each
recovery season. This may take several annual cycles of operation, each
showing an increase in recovery efficiency. Alternatively, a large initial
storage volume may be provided following construction. This may be
considered as the formation of a buffer zone, analagous to initial filling of
a surface reservoir. Once the buffer zone is formed, or the surface reservoir
is filled, ASR operations at the ultimate recovery efficiency can proceed.
The recovery efficiency attainable will depend upon the hydraulic and
water quality characteristics at each site. While 100% recovery efficiency
is a reasonable target and is obtained in most cases of storage in brackish
aquifers, lower recovery efficiency may occur in some situations due to
technical constraints or regulatory restrictions designed to promote aquifer
recharge. An economic analysis will then indicate whether the lost value
of the water not recovered is more than offset by the value of the water
recovered when needed. Usually this is the case.
GROUNDWATER RECHARGE AND WELLS
ASRWell
(ASR-1)
I~ 30 ~I
Lffhologlc Lffhologlc Natural Gamma
Description Column Ray Log
<
0
LIMESTONE. very
pale orange to '--,
~~
whrre abundant
cora line structure,
hard and cavernous 50
!(>'
LIMESTONE. very ! ... ~
100
pole orange to
whiTe chalky, I
porous, dense, with
some shell shards
~r-,
150
~ f--
i<>
k:>
~~
SAND, white poorly 200
consolidated,
angular to ~
subrounded,
interbedded
limestone lenses .!
b
f~t's
250
SANDSTONE. light
300
olive moderately
consolidated,
--.p
subangular to 1--,
rounded
r--~
350
r-
~>
~
I->
388'
400
<;;~
SAND. yellowish to c""p.
light olive-gray
""
428' unconsolidated, fine
to very coarse
450
;;==I-
grained quartz, well
4" 25 Slot graded, round to "--
20/30 PVC Screen
lO' SS Sump
427' to 432'
Gravel
Pack
388' to 428' w:~hi~fe~~J~~~T 11...--
~
clay lenses below
358' to 430' 428 teet 500
!'='
TD = ±432' TD = ±428'
)
550
j
0 20 40 60 80
LEGEND
API Units
Limestone
Ia Silty Sand
Clay D Quartz ~and
and Gravel
Calcareous
Sandstone
Geochemical Issues
The most complex issues pertaining to storage zone selection are with
aquifers, or portions of aquifers, that offer geochemical challenges. This is
discussed in greater detail in Chapter 5, Geochemistry. However, one
solution is to design the ASR well to case out production intervals that
contribute severe geochemical problems, if this can be achieved without
losing much of the potential production capacity of the well.
Typically, the detailed information needed to make a reasonable judg-
ment regarding well design to avoid geochemical problems can only
follow coring, core analysis, and geophysical logging. In the absence of
this data, it is difficult to know which intervals are contributing the water
quality constituents of concern. Consequently, the design of the second
and subsequent ASR wells may benefit from experience gained with the
first such well at any new site.
Screen Design
Pump Setting
One alternative that has not yet been applied in an ASR well is to set
the pump below the screen interval at the bottom of the well. For seasonal
operation, this practice may enable recovery rates well in excess of normal
continuous production rates. The specific yield of unconfined portions of
the aquifer will release much larger volumes of water from storage than
74 GROUNDWATER RECHARGE AND WELLS
Potentiometric
Confining Surface
...... ...... ..- . - - Layer
/
/
Long Term'-
Equilibrium /
Decline /
/
/
/
'\ I
'\ ; - - Short Term
\ 1 Seasonal
'\ Decline
Aquifer
Materials of Construction
Trickle Flows
Sampling Taps
Cascading Control
This is one of the more important elements of ASR wellhead design and
requires some care. Cascading occurs when the water level in the recharge
piping does not rise to ground surface during recharge. Allowing water to
cascade down the well can lead to significant plugging problems due to air
binding in the storage zone, and induced geochemical or bacterial activity.
Air present in recovered water can cause consumer complaints. Cascading
can also cause structural problems due to cavitation damage to pipes,
valves, and fittings. Cascading needs to be controlled in order to avoid
these problems, each of which causes plugging of the ASR well. Plugging
usually can be reversed; however, it requires considerable time and effort.
Water can be introduced into a well through the pump column, the
annulus between the pump column and the casing, one or more injection
tubes inside the casing, or some combination of these approaches. It can
be introduced under pressure or under vacuum, and it can be controlled
from the wellhead or from the bottom of the injection piping. Selection
among these alternatives is based upon consideration of several factors,
principal among which are the following:
• casing diameter
• static water level in the well
• type, size and capacity of the pump
• specific capacity and specific injectivity of the well
• expected production rate and range of injection rates
Some of this information may not be available at the time the design is
completed, creating the need for a flexible design approach, capable of
accommodating a reasonable range of expected conditions. It is usually
wiser to construct and test the ASR well to determine hydraulic perfor-
mance characteristics before finalizing design of the wellhead facilities.
This requires more time; however, it leads to better results. Provision of
flexibility is still advisable, since recharge rates can sometimes drop below
planned rates, causing unplanned cascading.
Annulus Recharge
High flow rates can sometimes be recharged down the annulus of a well.
To maintain positive pressure at the wellhead and thereby prevent cascad-
80 GROUNDWATER RECHARGE AND WELLS
ing, it is necessary to ensure that sufficient flows are always available for
recharge. When recharge flows fall below this critical rate, cascading will
occur and a vacuum will develop in the annulus and wellhead piping. Air
will be drawn into any open air relief or vacuum breaker valves, any leaks
in the upper portion of the casing or pump column, or elsewhere in the
wellhead assembly and will be carried down the well into the formation,
where it will tend to plug the well. This can happen due to reduction in
recharge flow rate or due to local or regional lowering of the static water
level in the storage zone.
A flexible solution is to seal the annulus at the wellhead and to ensure
that any wellhead valves that are connected to the annulus are closed
during recharge. In this way recharge can occur regardless of wellhead
pressure or vacuum, and under a full range of recharge rates, thereby
maximizing recharge volumes and reducing operating requirements.
A disadvantage of this approach is that water flows over a substantial
surface area of casing that is alternately wetted and dried. Therefore,
annulus recharge in steel casings has a high potential for production of rust
that can contribute to plugging the well during recharge and create regu-
latory problems during backflushing and initial stages of recovery. For
new wells, this can be avoided by selection of a non-ferrous casing or
coating where possible. For existing wells, particularly those with long
steel casings into low or moderate permeability aquifers, annulus recharge
can contribute to particulate plugging problems.
A second disadvantage of this approach applies, in particular, to retro-
fitting of existing wells for ASR purposes. Where the quality of well
construction is unknown or suspected to be poor, it is possible that the
casing or pump base may not be sealed adequately. Recharge would
therefore entrain air even if the wellhead piping and control valves were
sealed and closed, respectively. This can be checked by installing a tem-
porary packer in the well and pressure testing the casing to determine if it
will hold a given pressure for 30 min. This is sometimes referred to as
"mechanical integrity testing." Alternatively, a brief recharge test can be
conducted at a low rate in the supposedly sealed annulus, sufficient to
create a wellhead vacuum. Recharge is then shut off, and the vacuum is
monitored to see if it will hold for 30 min.
A related issue is that pressure surges have been known to occur in the
recharge piping of some ASR wells. In other ASR wells, recharge occurs
at higher pressures anyway, to overcome high static water levels or to
overcome density differences in saline aquifers. In such situations, the
pump bases should be designed to withstand expected operating and
transient pressures without leaking at the connection to the casing. A
DESIGN OF ASR SYSTEMS 81
flanged connection between the top of casing and the pump base, ma-
chined to ensure flat, parallel surfaces, and sometimes provided with a
circular groove and o-ring, can provide the required degree of sealing.
Recharge down the annulus of wells equipped with submersible pumps
requires care to ensure that the electrical cable port in the wellhead flange
is adequately sealed to prevent air entry during vacuum recharge or to
prevent leakage during pressure recharge.
Several variations on this annular recharge approach are possible. Re-
charge could occur down the annulus at sufficiently low velocity below the
water level in the well that any entrained air has the opportunity to bubble
out before reaching the formation. No known existing or planned ASR
sites utilize this approach, but in theory it should work. A downhole water
velocity in the casing below the pump would have to be less than the air-
bubble rise rate, or about 0.3 to 0.4 m/sec (1 to 1.3 ft/sec) for air bubbles
with diameters of 0.1 to 10 mm.
Another variation is to cease recharge at such times as cascading begins,
whether due to static water level decline or due to reduction in recharge
flows. This assumes that previous testing has shown that cascading causes
air entrainment in the well. Such an approach requires a degree of operat-
ing attention that is frequently not available. Larger ASR wellfields with
computer-controlled operations and telemetered monitoring parameters
can build this into their control systems; however, smaller systems are
more likely to continue recharge regardless of whether cascading is occur-
ring or not, with resultant reduction in recharge rates, due to plugging.
Methods to increase friction loss in the annulus have occasionally been
considered or tried. These have included sizing the pump to minimize the
annular space between the pump and the casing; addition of flanges at the
couplings in the pump column; inflation of a packer above the pump in the
annulus through an air or water line from the surface; and other novel
approaches such as adding floating objects in the annulus. Except for the
inflatable packer, each of these approaches has the same drawback in that
it is sometimes difficult to place in the well, or retrieve from the well, a
tightly fitting object. One that is not tightly fitting will probably not
provide much resistance to flow. Well casings are not always straight,
plumb, or round.
Several ASR sites utilize annulus recharge. Among these are Cocoa,
Peace River, Marathon, and Chesapeake. Cocoa and Marathon utilize PVC
casings; Peace River has epoxy-coated steel casings on all but the first two
ASR wells, and Chesapeake also utilizes an epoxy-coated steel casing. The
first two of these sites store drinking water in brackish aquifers within
which the depth to static water level is about 3 and -7 m (10 and-22ft)
82 GROUNDWATER RECHARGE AND WELLS
available. Above or below the flow range, the system will perform improp-
erly by either rejecting available flow or cascading.
Injection tubes are most applicable in situations where adequate opera-
tional attention is available to monitor and control flows at the wellhead.
They are also quite applicable in multiple ASR well systems where opera-
tions are controlled by telemetry and recharge flow variations can be met
by adding or deleting ASR wells from operation. In this way the relatively
narrow flow range of individual wells is not a substantial constraint.
Such an approach may require careful development of an operating plan
fGr those situations where the storage zone is brackish or otherwise con-
ASR wells. Alternate approaches include the use of food grade oil for
lubrication, or making special precautions to avoid vacuum development
in the pump column. Use of new column pipe is also advisable to minimize
the likelihood of leaks at the threaded connections.
For vertical turbine pumps, cast iron discharge heads are available in
various sizes, and fabricated steel heads can be made to accommodate
most configurations. Standard cast iron discharge heads can be machine
surfaced to fit a steel sole plate grooved for an 0-ring. This approach may
be useful for retrofitting an existing well for ASR purposes. It also pro-
vides a reasonable pressure seal for situations where recharge water levels
may rise above land surface during continued ASR operations or pressure
surges.
Examples of recharge through the columns of vertical turbine pumps
include Goleta Water District, CA; Calleguas Municipal Water District,
CA, Las Vegas, NV and one of the ASR wells at Kerrville. All four sites
utilize existing wells retrofitted for ASR purposes.
The type of well head seal will depend upon the type of pump in the
well. For submersible pumps, a flanged surface plate should be used.
Alternatively, a blind flange bored and welded to the column pipe can be
fitted to a ring flange on the well casing.
Combinations
All ASR wells experience a greater degree of water level change than
typical production or injection wells. This change in water level results in
air being drawn into, or released from the well during different phases of
operation. Adequate venting on the casing and on the wellhead discharge
piping should be provided in the form of air/vacuum release valves or
other form of vented opening. However, it is essential that these valves be
closed during recharge to prevent entry of air during potential vacuum
recharge. This is an important operating requirement, the omission of
which can entrain substantial quantities of air and plug the well.
Air relief valves are usually designed to vent air under relatively high
operating pressures. ASR wells usually recharge under much lower oper-
ating pressures at the wellhead. Sometimes under these lower pressures the
air relief valves will leak slightly. Provision for drainage of this leakage
water will avoid a problem that may be aesthetically unappealing (rust),
DESIGN OF ASR SYSTEMS 87
Pressure Gauges
Pressure gauges should be both durable and accurate. Sealed cases filled
with glycerin or silicone stand up well to harsh, outdoor conditions. The
fluid-filled gauges also provide needle damping if vibrations are present.
Pressure readings are useful in many places on ASR wellhead piping.
Consideration should be given to installing taps for pressure gauges at the
distribution system piping supplying the ASR well, upstream and down-
stream of any pressure control valves, upstream and downstream of any
wellhead filters, and on the wellhead recharge and recovery piping. If
vacuum or negative pressures may occur, particularly at the wellhead, a
combination vacuum/pressure gauge should be provided.
Gauges should provide the level of accuracy necessary for each loca-
tion. Generally, a gauge with 0.5% accuracy is desirable for the wellhead
but is not necessary at other locations.
To protect the gauges against damage during pressure surges, spikes, or
fluctuations, dampening devices can be installed for each gauge. These
range from a fitting provided by the gauge manufacturer to a simple, small
petcock.
88 GROUNDWATER RECHARGE AND WELLS
Flow Measurement
necessary to select a flowmeter type that will provide the desired accuracy
within the piping distance available. Straightening vanes are sometimes
used to straighten flow lines upstream and downstream of the meter.
For larger ASR systems, or those involving automated control systems,
it may be appropriate to obtain a certificate of proper flowmeter installa-
tion from the manufacturer.
For all ASR systems, consideration should be given to providing dual
flow measurement capability, at least for the duration of the test program.
Meter failure or loss of calibration during the test program has occurred at
several sites for a variety of reasons. Loss of calibration is difficult to
detect at the time, and usually only becomes apparent late in the program
when it is too late to repeat the tests. The resulting data can be difficult to
interpret. It is desirable to have two different types of flowmeters, one of
which is the primary meter. Any trend of increasing difference in measure-
ments between the two meters would signal the need for calibration or
meter replacement before proceeding further with the test program. These
problems appear to be more common with propeller meters that are used
widely in the water industry. Having a standby propeller meter or replace-
ment parts on hand can be helpful, available for rapid substitution if
necessary. A venturi tube or similar device incorporated in the wellhead
piping can provide the backup flow measurement during testing, and can
easily be removed when the system changes into long-term operation, if
desired.
Bi-directional flowmeters have been used at some ASR sites where it
was desired to convey both recharge and recovery flows through the same
pipe. However, bi-directional propeller meters have proven much less
reliable than corresponding venturi or magnetic meters.
Flowmeters utilized on ASR systems should include totalizing mea-
surement in order to monitor cumulative volumes during both recharge
and recovery. This is typically provided with propeller type flowmeters,
which are readily available, relatively inexpensive, and have been used
widely on ASR projects. Propeller meters are usually accurate to within
±2% of the actual flow rate.
Turbine meters are similar to propeller meters; however, they use a
turbine instead of a propeller. The turbine spins at a higher velocity and
subsequently requires a more precise bearing and mechanism. For this
reason, turbine meters are more sensitive to sand and particles in the water
flow. Upstream screens should be installed with turbine meters. They
typically provide higher accuracy and a wider operating range than does
a propeller meter. Typical accuracy is about ±1.5% of actual flow. Cost is
usually about 30% greater than the corresponding propeller meter.
DESIGN OF ASR SYSTEMS 91
vision for storing and handling the chlorine or other disinfectant that will
be used.
Chlorine gas added to water will typically result in a decrease in pH.
The magnitude of the decrease will depend upon the chlorine dosage and
the alkalinity of the water. Where the recovered water will be blended with
a much larger flow of water, the effect may be negligible. However, where
little or no blending will occur prior to consumption, the pH drop follow-
ing chlorination can be sufficient to produce an aggressive water, capable
of causing corrosion of pipes and fittings, and associated "red water"
complaints from consumers. The need for pH adjustment following recov-
ery is usually determined following construction and initial testing of ASR
facilities. Consequently, it is desirable to equip ASR wellhead facilities
with locations for chemical addition, if later required.
Adjustment of pH may also be advisable for recharge flows. Where
manganese is present in the storage zone, recharge at pH of less than about
8.0 may tend to cause the manganese to go into solution during an
extended storage period. Recovery of the stored water may then create a
problem with excessive concentrations of manganese and associated black
discoloration of wetted surfaces. Adjustment of the recharge water pH to
levels of about 8.5 or above will help protect against recovery of water
with high manganese concentrations.
Depending upon the potential for formation of disinfection byproducts
such as trihalomethanes and haloacetic acids when the recovered water is
disinfected, it may be necessary to add ammonia to the recovered water to
form a chloramine residual. Where ammonia is present in the recharge
water, its presence in the recovered water should be tested before making
a determination as to whether re-ammoniation is necessary. Typically,
ammonia is substantially reduced during aquifer storage. Reduction of
disinfection byproducts during ASR storage is discussed in greater detail
in Section 4.5, Disinfection Byproduct Reduction.
Pump Considerations
and head. The additional column pipe provides operating flexibility, since
the range of pumping water levels is usually not known until after a few
cycles of operation. Net positive suction head (NPSH) and motor electrical
horsepower should also be sufficient to match the full range of expected
pumping water levels.
Pump setting has been within the casing, or within a blank section
between screen intervals, in all ASR wells to date. However, it is antici-
pated that some future ASR installations may set the pump below the
producing interval in a bottom section of casing that serves as a sump. In
this way, seasonal production may be conducted at rates higher than those
associated with normal well operation, causing rapid seasonal lowering of
water levels and potential partial dewatering of confined or semi-confined
aquifers.
The volume of water available from dewatering a confined aquifer is
defined by the specific yield, which typically ranges from 5 to 35% of the
volume of aquifer material dewatered. In contrast, the volume of water
released from lowering of water levels within a range above the top of the
same confined aquifer would be defined by the storativity, which typically
ranges from 0.1 to 0.001%. Hence, in situations where hydrogeological,
geochemical, and bacteriologic considerations permit, it may be very
desirable to better utilize the large volume of water stored in an aquifer by
producing at a high rate for a short period of a few weeks or months. The
aquifer would then be recharged during the following low demand season.
This is illustrated in Figure 3.2. The pump design would then entail
additional column pipe and, for submersible pumps, a shroud around the
pump and motor to ensure that water flows around the motor during pump
operation, to provide adequate cooling.
To date, ASR wells have been equipped with vertical turbine, submers-
ible, and horizontal centrifugal pumps. All have proven adequate for their
specific applications.
In a few situations where storage zone permeability is very low, plug-
ging potential is deemed to be high, or discharge of initially turbid water
is a significant concern, consideration should be given to coating the
column pipe, both inside and outside, in order to reduce the surface area
subject to rusting during alternate wetting and drying periods associated
with recharge and recovery.
Normally it is wise to utilize the same pump manufacturer utilized for
other wells and pumping installations operated by the owner of the ASR
well. However, certain submersible pump manufacturers have indicated
that they will not honor the pump warranty if the pump is used for
injection. In this case, alternate manufacturers or types of pumps should be
94 GROUNDWATER RECHARGE AND WELLS
supply to the pump shaft at the beginning of each recharge period and
opening it at the beginning of each recovery period can also work; how-
ever, this is somewhat risky as a long-term operational requirement since
it would be easy to overlook the adjustment.
To date, no ASR wells have been provided with variable frequency
drives, providing for adjustment of recovery rates over a wide range.
However, some wells in California have been provided with two-speed
motors to facilitate energy recovery during recharge. One site in the
planning stages in southeast Florida is considering a two-speed motor to
enable recovery at normal rates to meet distribution system diurnal varia-
tions in demand, and higher rates if needed to meet fire flow requirements.
At this site, the ASR system is under consideration as a cost-effective
alternative to a new above-ground storage tank in an existing residential
area. Residents of this area oppose construction of the above-ground tank.
Pressure control valves may be required on either the recharge line, the
recovery line, or both. This provides operational flexibility in situations
where recharge pressures may fluctuate, where available flow may be
limited during certain hours of the day or months during the year, or where
recovered flows may interfere with system head curves at certain times.
A permanent survey benchmark should be provided at each ASR site,
showing the elevation. This will provide a reference point for measure-
ment of water levels.
ASR projects typically require substantial onsite testing during both day
and night. Consequently, it is important to provide adequate lighting, not
only for the ASR well but also for any observation wells that will be
measured or sampled at night. Electrical outlets at the site facilitate use of
test equipment, power tools, and other activities, the need for which may
not be apparent at the time the wellhead is designed.
If observation wells are to be sampled, consider how the samples will
be taken and if dedicated pumps should be installed in these wells.
Adequate site access is important. Delivery of chlorine cylinders and
suitable access for pump trucks is important. Adequate road access should
be provided so that cars can get to the wellhead, rather than just four-wheel
drive or track vehicles.
Provision of telemetry control is frequently desirable, particularly with
larger ASR systems, in order to reduce operational labor requirements. Not
only does this facilitate routine operations, it can also simplify data collec-
tion, monitoring, and reporting requirements. ASR systems may be changed
96 GROUNDWATER RECHARGE AND WELLS
• pump on-off
• pump failure alarm
• recharge pressure control valve setting
• recovery pressure control valve setting
• water level in ASR and observation wells
• chlorine residual
• recovery flow rate
• butterfly valve operation
• conductivity probe
• turbidity probe
Energy Recovery
3.3 WELLFIELDS
Dispersive Mixing
I ·: • I
Recovery Efficiency: (%) Recovery Efficiency: (%) Recovery Efficiency: (%)
(A) 80.0 (B) 82.0 (A) 74.0 (B) 82.5
• • ••••• ••
in all wells
(B) Recharge sequential,
center well first
• •
Recovery Efficiency: (%)
••
Recovery Efficiency: (%) (modified from MA. Merritt
USGS WSP 2261, 1985)
(B) 81.4 (8) 81.4
Advective Mixing
Some ASR wellfields may potentially store water in aquifers for which
the background advective rate of movement is significant relative to the
radius of the stored water during a typical ASR cycle. For example, the
cycle may entail water storage for several years to bridge drought/flood
periods or to meet emergencies. Alternatively, the storage zone may be an
unconfined aquifer, which typically has a greater rate of groundwater
movement than does a confined aquifer. In these situations, improved
recovery efficiency should be possible by elongating the wellfield design
in the direction of expected regional groundwater flow and providing for
a greater portion of recharge in upgradient wells and a greater portion of
recovery in downgradient wells.
Figure 2.8 shows an example of this kind of situation. It is a conceptual
layout of an ASR wellfield to store drinking water in a brackish, confined
limestone aquifer in Kuwait; it is designed to help meet seasonal peak
demands during summer months, and also to provide a strategic water
reserve for emergency purposes. The regional gradient would not be a
significant factor affecting recovery efficiency for annual ASR cycles;
however, that portion of the potable supply in long-term storage to meet
emergency needs would be subject to advective losses. Hence, the wellfield
is arranged in a linear fashion in the direction of regional groundwater
flow.
CHAPTER
Seleeted
ASR
Teehnieal
Issues
And God said, Let the waters under the heaven he gathered together unto one
place and let the dry land appear; and it was so. And God called the dry land
earth and the gathering together of' the waters He called the seas: and God saw
that it was good.
Genesis 1:9-10
ASR is not high technology requiring skills beyond the capability of all
but a few specialists in the field. But neither is it low technology. It is
somewhere in the middle of this range. The body of knowledge that
differentiates ASR from other water management and recharge technolo-
gies has been developed since about 1970 through investigations and
operating experience at several sites. Design considerations were dis-
cussed in Chapter 3, for wells, wellheads, and ASR wellfields. In this
chapter, several key technical issues are discussed in greater detail to
provide a broader understanding of the technology. Geochemistry issues
pertaining to ASR systems are discussed in Chapter 5. Taken together, the
information presented in these three chapters comprises the current status
of ASR technology development. Chapter 8 presents some probable future
directions for ASR technology.
101
102 GROUNDWATER RECHARGE AND WELLS
Definition
Assume for the example above that the average recharge TDS concentration is
200 mg/L; background TDS concentration in the aquifer is 1000 mg/L; the
drinking water standard for TDS is 500 mg/L; and during recovery, TDS
concentration increases as shown on Figure 4.1, reaching the target criterion
of 500 mg/L TDS at 80% recovery.
SELECTED ASR TECHNICAL ISSUES 103
Recharge Water
Quality - 200 (mg/1)
OL---------~--------~---------L--------~--------~
0 20 40 60 80 100
"1. Recovery
system, where blending can occur between recovered water and water
flowing through the plant or distribution piping. So long as the water
quality of the blend meets applicable drinking water standards, regulatory
criteria are met. Consequently, it is usually not necessary to terminate
recovery when drinking water standards are reached. Recovery can con-
tinue until such higher concentration is reached that the blend going to the
consumer approaches but does not exceed applicable standards. Obvi-
ously the target water quality criteria will depend upon a number of
factors such as the available blend with water from other sources during
recovery periods, water quality for these other sources, and local regu-
latory constraints.
For the situation where the ASR well is located within the distribution
system and consumers may receive ASR recovered water directly without
any blending, then drinking water standards will govern the target water
quality criterion. This is uncommon, based upon experience to date.
Na~ver-----------------------------------------------~
during off-peak months and therefore has relatively low marginal costs,
reflecting only electrical power, chemicals, and a small amount of opera-
tion and maintenance. The investment may be made over a period of
several years through successive full scale cycles in which increasing
volumes of water are recovered each year. Alternatively, it may be made
up front during several months of continuous recharge and no recovery.
The value of the buffer zone water invested is invariably quite small
relative to the savings achieved by proceeding with an ASR solution to
water supply needs.
The ultimate recovery efficiency attainable at any site has to be deter-
mined through testing and operations. At most ASR sites, 100% recovery
efficiency is attainable; however, the number of cycles of operation to
achieve this level may vary, as may the volume of buffer zone water
invested. Where 100% recovery efficiency is not attained after several
cycles, several factors may contribute to this result:
TABLE 4.1
ASR RECOVERY EFFICIENCY IN BRACKISH AQUIFERS
With
Trickle
Flow
-~
v
u:::
~50
Ljw;,~:~t~I. ·~-~~~~~~~~~~
Trickle Flow
Cycles 1-10
~
0
& LEGEND
Na~ver----------------------------------------------.
Recharge '--""--"'--4.---------:::---------------'
0 50 100
Recovery (%)
.I ................
\ Bacterial Growth
Time
Plugging Processes
Cake Filtration
Blocking with
Filtration Cake or Gel Filtration Compression
Total Volume
Suspended Solids
they reach the borehole wall, thus reducing the long-term plugging rate by
acting as a coarse pre-filter. It is possible that blocking filtration in the
filter pack continues while caking filtration progresses at the borehole
wall.
The next stage of plugging is cake or gel filtration. Cake filtration
begins when the layer of filtrate on the filter begins to thicken. The
resistance is directly proportional to the thickness of the filtrate. Cake
filtration in an ASR well is evidenced by a linear increase of injection head
over time while maintaining a constant injection rate. This linear response
conforms with the response of a membrane filter during the caking stage
of filtration.
Cake filtration continues until the filtrate thickness increases enough to
allow compression of the filtrate, thus initiating the final stage of plugging:
cake filtration with compression. Cake filtration with compression is
characterized by a sharp increase of resistance to flow, which is dependent
on the compressibility of the suspended solids. If this stage of plugging
occurs at an ASR well, continuing injection after this point may not be
practical due to the associated high plugging rate and/or resulting in-
creased difficulty of redeveloping the well. Identifying the beginning of
this stage of plugging during recharge may provide the signal for redevel-
opment of the well.
Suspended solids are present in the recharge water for virtually all ASR
wells constructed to date. While data on turbidity is readily available for
potable water sources, data on total suspended solids is not commonly
available. Experience at many different ASR sites has shown the presence
of an interesting range of solids in the recharge water, including sand, rust,
diatoms (single cell algae), alum floc, twigs, dead mice, live shrimp and
slugs. Accordingly it is wise to assume that solids are probably present and
take steps to quantify their occurrence and typical concentrations. This
provides a basis for remedial design and operational measures. Solids
typically occur in short intervals, probably associated with pressure tran-
sients and flow reversals in adjacent portions of the water distribution
system. Discrete, small volume samples are less likely to define the solids
loading in the recharge water than long-term composite sampling. Simi-
larly, samples from the bottom of a pipe are more likely to be representa-
tive than samples from the side.
Biological Growth
Geochemical Reactions
Particle Rearrangement
(&)
I!
where Lie!> norm = rate of plugging normalized for a recharge flow velocity
(flux) of 3 ft/hour at the borehole wall over a period of one year at a
temperature of 20°C; Li<P =rate of plugging (feet of head per year); qs =
standard flux (loading rate or velocity) at borehole wall of 3 ft/hour; q =
calculated average velocity (flux) at the borehole wall in ft/hour; this is the
injection rate/infiltration surface area (over the effective saturated thick-
ness or perforated/screened interval); lls =viscosity at a standard tempera-
ture of 20°C (centipoise); !l =viscosity at temperature of injection water
(centipoise).
The membrane filter tests were used to develop an MFI for each water
source. The MFI is represented by the slope of the straight portion of the
plot of time/volume (t/V) vs. volume (V), on a linear scale. Because of the
small amount of water and the short times used in the test, the reporting
units for MFis are sec/L/L.
MFis, as determined by plotting, were normalized to standard condi-
tions so that MFis measured with different pressure and temperature
conditions could be compared. The standard conditions used were a pres-
sure drop of 30 psi and a temperature of 20°C. The following equation was
used to normalize the measured values to standard conditions:
where MFI = slope of the straight portion of the plot of individual values
(sec/LIL); Jl 20 = viscosity of water at standard temperature of 20°C
(centipoise); Jl = viscosity of water at measured temperature in °C
(centipoise); P =pressure drop across filter (psi).
Bypass filter test (BFTs) are conducted on the source of recharge water,
to measure the average concentration of suspended solids over periods of
time ranging from a few hours to a week or more. The source water is
directed through a 5-J-Lm, 10-inch-long, spun polyester cartridge filter at
pressures ranging from 5 to 30 psi. Cartridges with smaller pore sizes (0.45
Jlm or 1 Jlm) are available, but have higher costs and shorter life expect-
ancy due to rapid plugging. A t1owmeter similar to those used by utilities
for household water use is installed in the filter piping, to measure the
volume of t1ow through the filter at each site.
Analysis Method
The bypass filters are used to measure the suspended solids concentra-
tion in the recharge water, over extended periods of time. The filters are
dried and weighed to the nearest 0.1 gin the laboratory. The totalizer on
the t1owmeter is read prior to putting the filters into service. The filters are
operated during injection until the flow rate through the filter decreases to
about 25% of the initial flow rate. When a filter is taken out of service, the
flowmeter is read, and the spent filters are put in plastic bags and delivered
to the laboratory for drying and weighing. The polyester filter material
cannot withstand the 105°C temperature of the standard drying oven.
124 GROUNDWATER RECHARGE AND WELLS
Therefore, the filters can be placed on top of the ovens and dried for
several days. The difference in filter weights and the meter readings are
used to calculate the concentration of suspended solids in the recharge
water. Filters utilized in tests to date were operated between 2 to 20 days,
with 10 days as the average service life.
Data was collected during testing at nine ASR sites, including informa-
tion regarding treatment and conveyance of the water prior to recharge,
well construction, recharge rates, pumping rates during redevelopment,
hydrogeology, and aquifer parameters. Table 4.2 summarizes well con-
struction and hydrogeologic conditions, while Table 4.3 summarizes the
ASR well testing characteristics and general information about the source
water. General information regarding seven of the nine sites is included in
project descriptions included in Chapter 9, Selected Case Studies.
The water level data from the ASR test wells and, when available, the
data from nearby observation wells were used to estimate plugging rates.
TABlE 4.2 PlUGGiNG RATE SITE INVESTIGATIONS: WEll CONSTRUCTION AND HYDROGEOLOGY
Las Vegas Valley Water 20 360-980 3,246 Leaky confined, sand and 175 250,000 400
District, NV gravel basin-fill deposits
Well i1A
Calleguas Municipal 14 670-930 953 Confined, sand and gravel 220 140,000-70,000 250-500
Water District, CA marine deposits
Well97
City of Pasadena, CA 26 192-629 1,089 Confined, sand and gravel 300 200,000-300,000 1,000-1,500
Garfield Well (total160) 300,000
Seattle Water 16 285-335 293 Confined, sandy gravel 165 370,000 2,500
Department, WA outwash deposits
Metro Test Well
Sonoma County Water 12 and 16 400-800 1,466 Unconfined, sand 56 20,000 50-100
Agency, CA and gravel
Occidental Road Well
Tucson Water, AZ
Weii8-44B 16 141-480 1,324 Basin fill with interbedded 200 65,000 200
Well C-14B 16 260-600 1,424 Basin fill with interbedded 220 25,000 60
Well C-26A 10 and 12 128-480 775 Basin fill with interbedded 250 50,000 180
TABLE 4.3 PLUGGING RATE SITE INVESTIGATIONS: WEll HYDRAULIC CHARACTERISTICS
Testing Characteristics
Rate Rate Monitor
Source Water
of of Well
Name of Agency and Injection Pumping Wellhead (distance Raw Water
Injection Well(s) (G/min) (G/min) Design in feet) Source Treatment Process
Centennial Water and Sanitation 260 410 Pump column pipe with None Mclellan Reservoir Chemical addition,
District, CO downhole control valve flocculation, filtration
WeiiA-6 and chlorination
Las Vegas Valley Water District, NV 1,460 2,000 Pump column pipe 870 and Lake Mead Chemical addition,
Weii11A 1,200 (Colorado River) flocculation, filtration
and chlorination
Calleguas Municipal Water District, 605 680 Pump column pipe 80 CA state project Chemical addition,
CA water flocculation, filtration
Well97 and chlorination
City of Pasadena, CA 1,550 1,600 Pump column pipe and None CA state project Chemical addition,
Garfield Well well annulus water flocculation, filtration
and chlorination
Seattle Water Department, WA 700 800 Conductor pipes in the 70 Cedar River, diverted Chlorination and
Metro Test Well well annulus via intake screening addition of lime to
reduce corrosivity
Sonoma County Water Agency, 590 1,500 Pump column pipe 75 Russian River, Chlorination
CA pumped from
Occidental Road Well collector well
system
Tucson Water, AZ
Well B-448 1,210 {Bailed 3- and 4-inch } 149 { Pumped
Well C-148 1,220 to conductor 60 groundwater Chlorination }
Well C-26A 1,060 redevelop pipes 50
SELECTED ASR TECHNICAL ISSUES 127
Centennial Water and 10 N/A N/A N/A N/A 3 3.010 0.130 0.167
Sanitation District, CO
WeiiA-6
Calleguas Municipal Water 14 N/A N/A N/A N/A 11 2.400 0.100 0.386
District, CA
Well97
Sonoma County Water 18 Membrane filter would run Est. <0.1 1 - - 0.002
Agency, CA for hours without clogging
Occidental Road Well
Centennial Water and 10 N/A N/A N/A N/A 3 3.010 0.130 0.167
Sanitation District, CO
WeiiA-6
Calleguas Municipal Water 14 N/A N/A N/A N/A 11 2.400 0.100 0.386
District, CA
Well97
Sonoma County Water 18 Membrane filter would run Est. <0.1 1 - - 0.002
Agency, CA for hours without clogging
Occidental Road Well
Centennial Water, CO Surface water, 10 N/A 0.167 220.3 260 524 4.0 96.6 40
Well A-6 treated
Las Vegas Valley Surface water, 20 16.1 0.083 3.4 1,460 3,246 3.6 2.3 400
Water District, NV treated
Weii11A
North Los Posas Surface water, 14 N/A 0.386 113.5 605 953 5.1 33.9 375
Basin, CA treated
Well97
City of Pasadena, CA Surface water, 23 N/A 0.015 24.9 1,550 1,089 11.4 1.8 1,250
Garfield Well treated
Seattle Water, WA Surface water, 10 79.2 1.750 4.3 700 293 19.1 0.1 2,500
Metro Test Well disinfected
Sonoma County Surface water, 18 Est. <0.1 0.002 0.0 590 1,466 3.2 0.0 75
Water Agency, CA disinfected
Occidental Road Well
Tucson Water, AZ
Well B-448 Groundwater 20 0.7 0.060 1.4 1,210 1,324 7.3 0.2 200
Well C-148 disinfected 22 1.3 0.052 2.9 1,220 1,424 6.9 0.6 60
Well C-26A 23 0.9 0.043 3.2 1,060 775 11.0 0.3 180
(/)
m
r
m
96.6 0
-!
m
0
)>
lOO.O m~f:o::~~~~~~lf1Tf~f=:i~: ~~11~1m=:~~~~~1~~J~~t~W.~f-9.7.111~n~~~ ~~~~~~~:: ----"-" (/)
JJ
----- ·-t"''--m9-tL--h-·-·-;- ---------;·----·-----G.3-86· nfl·-----.
············}···········:·-··:···~-- ...:.... :.. :..:................;.........;....... ;... :.····:····;... {··{ .... ~: .......{..... {........:.....:.. -!
m
0
... I
z
0
)>
r
u;
~ (/)
c
~ m
(/)
:§
c
'E 0.1
0 :::::::::: :::~::::::::::::::::: ~::::: : =~=~=~ :::: ::::::::::::::::::: ::~::::::: :::: :;: ~::;:::~::: ~=:: ::::: :::::~:::. =~ ::: ~== ~:: ~=~:
• ----~ ~-~r- •• --.--~ -.-- ~ - -r--r--(--·- ------ ---·--- ---.--- -..---~- -.--:- "'~- " t - - - - - - - - __ "'! ___ - - - : - - - - . : - - - : - - '1'-~-
z
:::::: ::~~:: :::~:: ~~:: ::q:~:~: ·:::::: ::::~~~ ::r: :E: r ~: : : ~: r: ::::: ~et~ w~-.J
-----. ···c·· ---;·-· ·r·- . ·r·r-,· ·;· ------ •• -;-·-
··o·-- ";".-,-. c--.-o· ,. ---- ---1'71::.: -....... --.- "Jj';" •
............ f··· .. ······:· ... ~. . ·+··1· -;- -~- .............. -~- .. ·····1·"''' ·1·. ·i·. ·1·<···1··1· .... ······· ·+· -~ ·•·•:•91: ~-·
'
00 ' ' ' ' '
...
, 0.002 mg/L
~cldonl,. Rd.""'' ]
00
10 1.000 10.000
Aquifer hydraulic conductivily - gpd/ftl
Note: The size of type used to label the data points is intended to be roughly
representative of the magnitude of suspended solids in the source water.
Figure 4.7 Relationship between clogging rate, hydraulic conductivity and total suspended solids. ........w
132 GROUNDWATER RECHARGE AND WELLS
does not follow the pattern of other wells is possibly due to inability to
accurately measure the controlling factors or due to other factors that have
not been identified or adequately accounted for. The data from the Occi-
dental Well is possibly the most significant data presented in this study,
because it demonstrates conclusively that for low aquifer permeability and
low suspended solids content in the source water, plugging does not occur.
Well A-6 is another important data point, since it demonstrates that for low
aquifer permeability and moderate suspended solids content, the plugging
rates are high. The Metro Well demonstrates that low plugging rates can
occur with high suspended solids if the aquifer is highly permeable.
Data from additional ASR well sites would further define the relation-
ships presented here. However, this analysis provides a reasonable ap-
proach for estimating plugging rates at new well sites prior to well con-
struction and testing, based upon literature values for aquifer parameters,
assumed well design, and field measurements of recharge water character-
istics. Estimated plugging rates, in tum, can provide a basis for well
selection, design, and pretreatment to achieve acceptable backflushing and
redevelopment frequency and satisfactory operational performance of re-
charge facilities.
Following is a theoretical example of how this approach may be used
to guide ASR feasibility investigations at a potential new site.
area of about 470 ft 2. For consistency with the way in which the data in Figure
4.7 were generated, the screen area is utilized for normalizing the data.
Estimated recharge rate is about 0.5 MG/day, or half of the typical well
production rates in the area. Recovery specific capacity is estimated from a rule
of thumb (transmissivity/2000), or from local experience, at 10 G/min/ft.
Recharge specific capacity is expected to be about half of recovery specific
capacity.
From Figure 4.7, the normalized plugging rate is estimated at about 0.25 ft/
month. This should be multiplied by a factor of 1.2 to account for temperature
and viscosity differences. It should also be multiplied by a factor of 31 to
account for the flux rate difference at the borehole wall. Adjusting for temperature
and flux normalizing factors, the expected plugging rate is about 9.3 ft/month.
At the beginning of recharge, water level in the ASR well would be about
20 ft above land surface. During a typical recharge season, water levels may
rise due to plugging to 70ft above land surface within about five months. This
would effectively reduce recharge rates, since the pressure available from the
distribution system may be insufficient to overcome further plugging head
losses. The need for periodic redevelopment is indicated. A monthly frequency
would maintain water levels within an operating range of about 20 to 30 ft
above land surface and would be more likely to eliminate residual plugging that
can occur when heads build up so high that pumping the well is insufficient to
restore its recharge or recovery capacity.
Redevelopment
able water level decline in the ASR well during pumping. For the example
above, where the recharge specific capacity is estimated at half of the
recovery specific capacity, the recharge rate could initially be set at half
the recovery rate, or a slightly lower rate with anticipated less-frequent
backflushing. The actual plugging rate would then be monitored to com-
pare with the expected value. The well would then be pumped to waste for
a few minutes or hours to purge solids from the well. Assuming that
recharge and recovery specific capacity are restored, the recharge rate or
duration could be extended in small increments in later cycles, each of
which would show greater plugging. So long as redevelopment pumping
was able to restore specific capacity, the incremental increases in rate or
duration would continue until either the desired recharge rate had been
achieved, the duration extended to a full recharge season, or signs of
residual plugging became evident, such as inability to easily restore re-
charge specific capacity. The ideal situation is one in which the plugging
rate is sufficiently slow that redevelopment only needs to occur at the
beginning of scheduled recovery.
Redevelopment pumping or backflushing usually involves pumping the
ASR well to waste for anywhere from 10 min to 2 hours. Surging the well
by alternately turning the pump on and then shutting it off for two or three
cycles in a period of about 3 hours or less is practiced at some sites. This
is usually sufficient to restore specific capacity during recharge. However,
as discussed in Section 3.2, Design of Wellhead Facilities, care has to
exercised to avoid damaging the motor on submersible pumps or the shaft
on all pumps by restarting it too soon after shutdown.
The frequency of redevelopment pumping varies substantially between
ASR sites. Table 4.7 lists a number of operational ASR sites and the
typical redevelopment frequency. Information is also included regarding
w~--------------------------------------------------~
Cycle
Figure 4.8 ASR well increase in specific capacity, Manatee County, Florida.
storage period between recharge and recovery was on the order of an hour
or less.
A similar approach has been considered, but not implemented, for
storage of aggressive waters produced from desalination plants in the
Arabian Gulf. These plants typically are located over brackish limestone
aquifers, some of which would be suited to seasonal, long-term, and
emergency storage of drinking water while similarly achieving savings in
water treatment costs through pH stabilization in the aquifer.
commissioned. The same units have been in use for wastewater renovation
since the mid-1980s. Operating pressures for these units range from 25 to
40 psig, while the pressure differential across the membrane varies from
2 psig for a clean membrane to 15 psig when the membrane is fouled.
Particle sizes are reduced to below 0.2 J.lm with this process. The backwash
volume is about 2 to 7% of the feedwater volume. The membranes are
chemically cleaned when pressure differentials exceed about 15 psig,
using caustic-based solutions at pH values above 12. These units retain
protozoan cysts such as Giardia and Cryptosporidium, nearly all bacteria
of health concern, and turbidity. They also provide between 2- and 4-log
removal of viruses.
Installed manufacturer prices for Memcor units range from $1.00/G
installed capacity for a unit capable of treating 360,000 G/day; $0.50/G
installed capacity for a unit capable of treating 1.1 MG/day, and $0.35/G
installed capacity estimated for a unit capable of treating 23 MG/day.
An alternative approach is manufactured by Kalsep, Inc., called the
Kalsep Fibrotex System. As applied to alum-floculated surface water for
reverse osmosis pretreatment, this unit is estimated to cost about $300,000
to treat 2.5 MG/day of water with a TSS concentration of 1 mg/L. Result-
ant particle sizes in the product water are estimated within the range of 1
to 3 J.lm.
Considering the range of alternatives presented above, it appears that
sand filters, ring filters, drum filters and horizontal wells can filter re-
charge water to small particle sizes generally suitable for agricultural
applications that would not plug irrigation systems. In some cases, these
may also be suitable for ASR wellhead filtration, particularly with storage
zones that have high transmissivity. Where aquifers have lower transmis-
sivity, other filtration systems are available that can reduce particle sizes
down to between 2 and 10 J.lm using multi-media pressure filters or
cartridge filters. Microfiltration systems using membrane filter technology
can remove particles down to 0.2 J.lm. Selection of the appropriate technol-
ogy to meet the technical and regulatory requirements for ASR operation
has yet to be clearly defined.
It appears that agriculturally oriented systems can be supplied at costs
acceptable to the agricultural community, since such systems are already
in wide use. At higher manufacturer's costs of roughly $30,000 per 1 MG/
day system, wellhead filtration can be provided to reduce particle sizes to
levels that would probably be compatible with most ASR systems. Even
higher levels of treatment can be provided with membrane and comparable
processes, for unit costs in excess of $100,000 per MG/day installed
capacity.
140 GROUNDWATER RECHARGE AND WELLS
• THMs and HAAs are removed from chlorinated drinking water during
aquifer storage.
• HAA removal precedes THM removal; the more brominated species tend to
be eliminated earliest.
144 GROUNDWATER RECHARGE AND WELLS
• In most cases, removal of these halogenated DBPs does not appear to occur
until anoxic conditions develop, and frequently follows the onset of denitri-
fication. A biological mechanism is suggested. Additional work must be
conducted to establish the mechanism(s) responsible for removing these
DBPs, and the conditions under which they occur.
• THM and HAA precursors are also removed to a significant degree during
aquifer storage.
For the five sites investigated, THM and HAA reduction occurred
during a few weeks of storage, as opposed to days or months. Site-specific
testing is needed at other sites to establish reductions that may be achieved;
however, it appears that seasonal ASR storage can provide useful water
quality improvement benefits for many utilities. Whether or not THM
formation potential is reduced at any particular site during ASR storage,
the reduction in instantaneous THM concentrations may provide a signifi-
cant reduction in total THM concentrations (instantaneous THM plus
THM formation potential) once the recovered water is re-chlorinated prior
to distribution.
This is a significant benefit attributable to ASR systems. For some ASR
sites, it may be possible to recover water with low DBP concentrations,
blend it with higher DBP water from the treatment plant, and thereby meet
70
IJ TTHM
60 llll THAA
_.
~50
..... Note: Recovery began on Day 111 .
a"'c
0
:a:: 40
_g
&::
Ill
(.)
c 30
0
u
20
10
0
Rch1 Rch2 Rch3 3 22 50 78 112 119 126
Storage time (days)
Figure 4.9 Disinfection byproducts during ASR storage, Thames Water Utilities,
England.
(/)
m
r
30 m
0
-1
0 CHCI3 m
0
)>
IS:! CHBrCI2 (/)
25 :D
II CHBr2CI -1
m
~ CHBr3 0
I
z
0
)>
~ 20 r
0)
::;, Ui
....... NOTE: Recovery began on Day 111 . (/)
c: c
0 m
(/)
15 15
.::
c:
Ql
0
c:
8 wJI ~ ~~ ~ I ~SSM - 1'9'::
0
Rchl Rch2 Rch3 3 22 50 78 112 119 126
Storage time (days)
Figure 4.10 Trihalomethane constituent distribution during ASR storage, Thames Water Utilities, England. .....
ol>o
tn
.....
.j>.
(j)
140
0 THMFP
l.l HAAFP
120
0 DO
!Ill N-N03
5
82
0
Rch1 Rch2 Rch3 3 22 50 78 112 119 126
Storage lime (days)
Figure 4.12 Dissolved oxygen and nitrate concentrations during ASR storage,
Thames Water Utilities, England.
DBP standards that are expected to be lowered within the next few years.
Other potential benefits are the opportunity to recharge water with a
chlorine residual, avoid the addition of ammonia, and count on biological
processes in the aquifer to reduce resulting DBP concentrations prior to
recovery. As discussed in Section 4.2, Well Plugging and Redevelopment,
changing from chlorine to chloramine disinfection in Los Angeles, CA
140
~
,
r::-. ~ ........ - - THMFP
- - THM
•§"1oo ~ \ \
120 I( Kerrville, Texas
............ \
T Thames Water, England
........... \
........__ c Centennial WD, Colorado
"
las Vegas, Nevada
• 80
'\ ~
g' \ ~~ - r- p
... ...
~
... ...
K
140
- HAAFP
120 - - HAA
K Kerrville, Texas
& r Thames Water. England
c Centennial WD, Colorado
p Peace River, Florida
eg! 100
< 80 v Las Ve as, Nevada
~
_g 60
p
Ill G)
at 40 JJ
0 0
;!. 20 T c
z
0
0 ~
0 2 3 4 5 6 7 8 9 10 11 12 m
JJ
Time (weeks)
JJ
m
0
Figure 4.14 Haloacetic acid reduction during ASR storage at five sites. I
)>
JJ
G)
m
)>
z
0
~
m
r
r
(!)
(f)
m
r
m
~
m
0
)>
(f)
::Il
-i
m
0
TABLE 4.8 DISINFECTION BYPRODUCT REDUCTION DURING ASR STORAGE I
z
0
)>
Storage Expected r
Disinfection Byproduct Reduction (%)
Duration Reduction Due (jj
(f)
Site (days) to Dilution (%) TTHM TTHMFP HAA HAAFP c
m
(f)
Thames Water Utilities, England 111 64 100 >51 100 >76
Upper Guadalupe River Authority, Kerrville, Texas 127 50 25 24 100 (4)
Centennial Water and Sanitation District, Highlands 49 0 92 (<80) 100 12
Ranch, Colorado
Peace River/Manasota Regional Water, Supply 90 0 100 (-11) 100 100
Authority, Arcadia, Florida
Las Vegas Valley Water District, Las Vegas, 36 25 33 44 100 90
Nevada
...
""'<D
150 GROUNDWATER RECHARGE AND WELLS
injection wells has been shown to encourage bacterial activity around the
well, increasing both plugging rates and redevelopment difficulty.
Following completion of the field testing for the AWW A Research
Foundation program discussed above, preliminary laboratory investiga-
tions were conducted to evaluate the mechanisms governing the DBP
reactions that had been observed. Results were inconclusive, reflecting the
complexity of the laboratory procedures; the volatile nature of some of the
compounds; and the need to maintain anoxic, aerobic, or sterile conditions
in the sample vials. Further investigation is required to establish the
mechanisms responsible for the observed DBP reactions occurring in the
field.
was achieved with addition of sodium bicarbonate, while at the other site
sodium hydroxide was used. In general, the latter approach is favored,
primarily for operational reasons. Pretreatment at these two sites is re-
quired for all recharge flows.
Following is a brief discussion of pre- and post-treatment issues related
to several water quality constituents or processes pertinent to ASR opera-
tions. Particular emphasis is included for iron and manganese issues, due
to their prevalence in potential ASR storage zones.
Disinfection
Iron
0 . 8 . - - - - - - - - - - - - - - - - - - - - - - - - - -.....
~ 0.6
0)
g
c
~
~
~ 0.4
5
0
.2
c
0
E Recovery ________ - -
§
..... 0.2 ,, ..
,, "
\ I \. ,'
'--- I
._ "=> _ ,
',
'
, __ """--
~-'"
Figure 4.15 Ammonia concentrations observed during Cycle No. 5, Well R-1,
Port Malabar, Florida.
ASR facility. Both approaches are discussed below, along with other
measures for controlling iron concentrations and potential plugging with
ferric hydroxide.
Aquifer Pretreatment
After the first six test cycles at Swimming River, the pH of the recharge
water was further increased to 9.0 with addition of sodium hydroxide, to
address concerns regarding low residual concentrations of iron in the
recovered water. This change also addressed new concerns regarding
manganese concentrations in the recovered water, resulting from the acid
154 GROUNDWATER RECHARGE AND WELLS
14,-----------------------------------------------------.
10 Volume(MG)
~ Cycle Recharge Recovery
0::::::.
sc:
01
1
2
10
10
7
5
0
'5 8
3
4
9
10
8
7 , ,
J::
c
CD
5
6
52
3
47
3
, ,,
g 7 4 3 ,
,,
0
u 6
c
_g ---~Cycle 1
I
I
:e,!}, - · - Cycle2
-cycle3
I
4 •• •• • Cycle 4
=cycleS
... Cycle 6
•·••·•·•••·•·· Cycle 7
..
2
0 10 20 30 40 50 60 70 80 90 100
Percent Recharge Water Recovered
Vacuum Degassification
Post- Treatment
Manganese
Arsenic
No field work has yet been conducted regarding the use of ASR wells
for arsenic reduction. However, much attention has been directed in recent
years to the public health significance of arsenic in drinking water, and
new regulations are being considered by the U.S. Environmental Protec-
tion Agency that may reduce drinking water standards for arsenic.
In the near future it is anticipated that work will commence to establish
the effectiveness of ASR wells in removing arsenic during aquifer storage.
The reaction mechanism is expected to involve co-precipitation of arsenic
along with ferric hydroxide in aquifers containing low concentrations of
iron-bearing minerals and recharged with water containing dissolved oxy-
158 GROUNDWATER RECHARGE AND WELLS
0.4.-----------------------......-,,.-r-----,
~ 0.3-
sc:
----Cycle 1
-·-Cycle 2
0 - Cycle3
,
~
'E
•• •• • Cycle 4
=cycleS
,,
Gl
0
,"
• • • Cycle 6
""
c:
8 0.2- ············· Cycle 7
""
i
I
I
~
0 .. .·
~ ""J ........
~ 0.1-
100
Percent Recharge Water 11eeovered
Radon
lon Exchange
None of the currently operational ASR systems in the U.S. have expe-
rienced plugging or water quality problems due to ion exchange between
the recharged water and the native water in the aquifer. Geochemical
modeling has been performed for almost every site, sometimes indicating
the potential for ion exchange to occur. Water quality data collected at
several ASR sites have confirmed the occurrence of ion exchange reac-
tions, but not to an extent that would threaten ASR operations.
Perhaps the best-documented example of ion exchange plugging was at
Norfolk, VA, during a U.S. Geological Survey injection well recharge
investigation conducted in 1971-1972 [ 10]. The storage zone was a clayey
sand aquifer containing brackish water. Recharge was with treated drink-
ing water. During the first three test cycles, clay dispersion was shown to
cause uniform plugging of the storage zone throughout the whole length
of the screen interval, with 94% reduction in specific capacity of the well.
Subsequent redevelopment and treatment with calcium chloride as a pre-
flush prior to recharge Cycle 4 was successful in stabilizing specific
capacity of the well at about one third of the initial value prior to testing.
The authors concluded that "treatment of the injection well with a clay
stabilizer prior to injection of any freshwater will minimize clogging and
increase recovery of freshwater."
The pre-flush solution selected at Norfolk was calcium chloride, pro-
viding a divalent ion, calcium, to exchange with monovalent sodium ions
in the formation clays. A trivalent ion such as aluminum would also
achieve the same purpose.
Analysis of cores obtained from potential ASR storage zones, particu-
larly in unconsolidated, brackish aquifers, can provide an indication of
the potential severity of the clay dispersion problem. If smectite (mont-
morillonite) clays are present, even at concentrations of 1% or less, and
the aquifer contains brackish water, the need for pretreatment may be
indicated.
One element of the solution to clay dispersion problems may be to allow
time for a gradual adjustment in salinity of the aquifer around the well. A
slow transition may yield fewer plugging problems than a shock treatment.
Since the greatest opportunity for rapid salinity change occurs next to the
well screen and gravel pack, efforts to ensure a gradual adjustment in
160 GROUNDWATER RECHARGE AND WELLS
0.50
..._
:::::!'
C)
g
0.40
-8 Background Hydrogen Sulfide 3.5 mg/L
""
:;
"'c 0.30
ell
~ ....
'0
>-
:I:
0.20
.....
,. ..... ,/'
,...,.,.
_..
-- -
:,.··
0.10
./·· /
..-·· ~--·-
...... -~
__
~0 ...
0
0 10 20 30 40 50 60 70 80 90 100
%Recovery
salinity change should focus on initial recharge flows into the well. Low
initial flows, or alternate periods of no flow, may help to achieve this
objective.
While ion exchange may be reversible, the plugging associated with ion
exchange is frequently not fully reversible. This is because clay particles
that are dispersed through contact with water of low ionic strength tend to
move with the water until such time as they become trapped in the aquifer
matrix. Redevelopment may dislodge some of these particles and possibly
remove a few of them from the well; however, it will not restore the
particles to their original location.
Hydrogen Sulfide
tions of about 0.2 mg/L at the end of the first ASR cycle and decreasing
to lower concentrations during subsequent cycles.
• How can the ASR component of a water system be best utilized to ensure
the least-cost expansion path for water supply facilities to meet projected
demands?
• How should the ASR wellfield be operated to meet projected seasonal, long-
term, emergency or other needs with adequate reliability?
• How far will the storage bubble extend around the ASR well? How much
of the stored water can be recovered? What, if any, will be the effect of
biological reactions occurring underground upon recovered water quality?
• What geochemical reactions are expected to occur due to mixing between
stored and native water in the presence of aquifer clays and minerals? Are
these reactions significant and is the reaction rate significant to proposed
ASR operations?
This model has been developed by CH2M HILL, to simulate the opera-
tion of a water system comprising a river source of supply, intake struc-
ture, offstream reservoir, water treatment plant, and ASR system. It has
been applied to the Peace River ASR system, to determine the most cost-
effective facilities expansion path to meet projected demands with accept-
able reliability. The number of variables, and the interrelationship between
some of the variables, complicates any simple analysis of this system.
162 GROUNDWATER RECHARGE AND WELLS
Finished Water
Transmission
This model has been applied to the ASR system at Kerrville, to evaluate
whether use of ASR wells to maintain groundwater levels at a target
elevation above mean sea level would enable the Upper Guadelupe River
Authority to meet a repeat of the worst drought on record if it were to occur
in the future. Further discussion of the Kerrville ASR system is included
in Chapter 9, Selected Case Studies. The alternative water supply option
for this site was the construction of a new offstream reservoir at a cost at
least five times that of the ASR system. The operating rule for this site is
to recharge whenever water is available from the currently 19 megaliters/
day (5 MG/day) treatment plant after meeting system demands. Recharge
ceases when the elevation reaches 1500 ft in the ASR well. Water is
recovered to meet seasonal peak demands each year, leaving some water
underground as carryover storage to meet the drought demand at such time
as it may arrive.
The same model has also been applied to the Peace River ASR system
to estimate drawdown effects upon adjacent well owners due to seasonal
recharge and recovery operations. The ASR recovery capacity at this site
needs to equal maximum day demands so that these demands can be met
during times when there is no flow available from the river to treat at the
plant. Flow distribution to each well has to be balanced during both
recharge and recovery so that well interference does not push stored water
away from individual wells.
50
~
v
40 ~
/
~ 30
v
~
11::
.! (
I
,g 20
w
10
0
1/
0 10 20 30 40 50
Storage Volume (million gallons)
Figure 4.20 HST3D simulated relationship between recovery efficiency and initial
storage volume, Marathon, Florida.
SELECTED ASR TECHNICAL ISSUES 167
60 r----,----.,-----.,-----....,,....storage; 0 Days
-Storage; 30 Days
;::::-storage; 60 Days
Storage ; 90 Days
-Storage ; 180 Days
3 4 5
Cycle Number
during successive recharge, storage, and recovery cycles. This model has
been upgraded by CH2M HILL to make it more user-friendly and WIN-
DOWS-driven. Once calibrated against existing data sets from operational
ASR sites, it should prove useful as a predictive tool at proposed ASR sites
to estimate recovery water quality and recovery efficiency.
Several other solute transport models are available that could be applied
to ASR systems. Among these are RESSQU, KONIKOW -BREDERHOEFT,
and SUTRA.
CHAPTER
Geoehemistry
The world appears as you perceive it. It is not that your perceptions are wholly
shaped by a so-called objective world. The habit of interpretation is interactive;
we do things to test our hypotheses until we have created a complex web of
sensory input and centrijitgal manipulation. By the time we are "mature," we
have created innumerable layers of intClpretation and biased perception that
become the templatesj(Jr living. Of course, we could have fun with this situation.
We could change the templates that we use to interact with the world.
Deng Ming-Dao, 1992, 365 TAO: Harper San Francisco
5.1 INTRODUCTION
169
170 GROUNDWATER RECHARGE AND WELLS
flaws, the approximate severity and impact of such fatal flaws, and the
potential solutions can provide a useful guide to those considering ASR
projects.
In areas with proven, long-term satisfactory performance of ASR sys-
tems and with no signs of geochemical problems, it is probably reasonable
to assume that such problems should not occur in future projects. For
example, this is the case in Florida ASR systems storing treated drinking
water in brackish limestone artesian aquifers. It is also believed to be the
case in Las Vegas, NV, where the storage zone is a fresh, unconfined
alluvial aquifer.
In areas with little or no prior ASR or recharge well performance to use
as a guide, it is wise to consider potential geochemical issues carefully and,
based upon geochemical and other test results, adjust program develop-
ment to reflect needed changes. It is important that test program develop-
ment should reflect an awareness that ASR geochemistry is not yet an
exact science. A potential problem that is indicated by initial laboratory
analysis, core testing, and geochemical modeling may not materialize, or
may occur so slowly as to be of no real significance. On the other hand,
an unexpected geochemical problem may arise, requiring procedural
changes, despite reasonable efforts to identify such problems in advance.
With steadily increasing experience, the occurrence and frequency of
unexpected problems should decline. This chapter presents experience
through 1993, recognizing that the ASR geochemistry field is advancing
fairly rapidly.
The outline of this chapter generally follows the geochemical steps that
are considered appropriate for ASR feasibility investigations, starting with
sampling of recharge water and native groundwater, laboratory analysis,
and a preliminary geochemical assessment. This is frequently followed by
observation, well construction and coring, core analysis, and more detailed
geochemical evaluation based upon computer simulation of the mixing of
recharge water and groundwater in the presence of the aquifer clays and
minerals. This completes the procedure for sites that are considered
geochemically benign. However, where uncertainty still exists regarding
geochemical effects, column testing or laboratory batch testing may be
performed to simulate in the laboratory the operations planned in the
wellfield. It is less expensive to destroy a core than it is to irreversibly plug
a well through geochemical reactions. This laboratory testing provides the
basis for pretreatment or post-treatment plans that may be required during
ASR well testing and subsequent operations.
GEOCHEMISTRY 171
Parameters
At least one sample of the recharge water is required for the geochemi-
cal analysis, taken at a time of the year that is representative of expected
recharge water quality. In addition, a representative sample of the ground-
water from the proposed ASR storage zone at or adjacent to the selected
test well site is required prior to any recharge. These samples are analyzed
for parameters listed in Table 5.1. Some of the analyses are performed in
the field and the remainder in the laboratory.
Temperature, pH, and specific conductance are easily measured and
need to be determined in the field. The pH of groundwater, and to a lesser
extent surface water, typically changes as much as a full pH unit between
the collection time of a sample in the field and the time it is measured in
Parameter
pH
the need to filter and acidify the sample in the field if reliably significant
geochemical interpretations or decisions are to be made with the data.
Table 5.3 [2] shows the common, high, and low ranges of pH values for
natural waters. The high range generally requires a significant bacterio-
logical activity to achieve under natural conditions but alkaline, magne-
sium-silicate rocks can reach these pH values. Grout failure should also be
considered since it can also cause high pH values. Low pH ranges are
probably bacterially-catalyzed oxidation of primarily iron sulfides (pyrite,
pyrrhotite, and marcasite). A pH of -2 has been reported in a mine but
should be very rare under natural groundwater conditions sinc,e rocks
surrounding the iron sulfides buffer the pH. Furthermore, the iron sulfides
build up an oxidized iron hydroxide coating as a result of continued
exposure to groundwater. This coating protects the sulfide, minimizing
oxidation.
Specific Conductance
The ratio between TDS and the specific conductance should be between
approximately 0.55 and 0.75 for most natural waters. A value of 0.67 is a
common ratio for natural waters with less than about 1,000 mg/L TDS.
This is the first check on the TDS concentration.
Mass Balance
The error should be within 5% for a high quality analysis. A 10% error
is allowable if TDS is less than approximately 100 or more than 5000 mg/
L. An error equal to or greater than 20% is not acceptable for meaningful
interpretation of aquifer geochemistry. However, water containing a high
iron concentration can give a significant cation error, and a high nitrate
concentration can cause a significant anion error if not included in the
ionic balance. A very high pH water will not balance unless the hydroxide
is added.
A comparison between the calculated meq/L of either the cation sum or
anion sum should be approximately equal to the specific conductance
(1-lmhos/cm at 25°C) divided by 100. These are typically within 5%,
assuming that the major cations and anions have been analyzed and the
TDS is approximately 1,000 mg/L or less. The cation or anion sum closer
to the value of the specific conductance divided by 100 is probably correct,
and the sum farthest from the value probably contains any error. The
laboratory should be consulted and questioned about any potential error.
These are two of several methods for showing water chemistry types
graphically. They show patterns, allowing the eye to discriminate water
chemistry types. The size of the individual diagrams indicates the relative
amount of TDS because the diagrams plot each cation and anion in
milliequivalents. Figures 5.1 and 5..2 illustrate these two types of dia-
grams. These diagrams are best used for illustration of water chemistry on
maps.
Na + K 1--t----l----l'---1----1----1 Cl
Ca HC03
Mg so4
Fe co3
30 25 20 15 10 5 0 5 10 15 20 25
Cations Anions
(milliequivalents per liter) (milliequivalents per liter)
10
17-3
Na+ K
CabMg 12-6
so/T'~
Cl HC03
Cl
Ca M 15-1
so~H~oCl 3
0 5 10
Hydrogeochemical Processes
I
so4
+CI
I
Mg
0 L,~~~~~~~~~~~~ L-~~~L2~~~L-~~~~I 0
100 100
Cations -co Cl- Anions
Percentage Reacting Values
Oil Field
Brines (OFB)
SGW
"-
Oo '() I
I vox
o o • I
-cA CL-
4.9 in this case. Weathering alkaline rocks, like volcanics, will increase the
pH to above 8.0 and yield a sodium-bicarbonate water type.
Biomass production assimilates ammonium, sulfate, calcium, magne-
sium, potassium, iron, and aluminum into the bodies and fluids of micro-
organisms, thereby converting part of the ammonium to nitrate and pro-
ducing organic acids that lower the pH less than occurs with nitrification.
Denitrification, on the other hand, is conversion of nitrate and nitrite to
nitrogen gas, and is an anaerobic process that occurs under reducing
conditions. Most reduction processes are hydroxide, or alkaline, producing
processes. In this case, the pH increases to about 4.7.
Sulfate reduction occurs under relatively severe reducing conditions,
typically below an Eh of -100 mV, created mostly by the microorganisms
responsible for converting dissolved sulfate into either hydrogen sulfide
gas (rotten egg odor, pH less than 7 .0) or bisulfide ion (HS-, no odor,
potential precipitation of iron, pH equal to or greater than 7.0). Actually,
the microorganisms produce organic-sulfur compounds, thio-species, which
can break down into one of the above inorganic species depending upon
transport into an oxidizing environment (acid-producing, hydrogen sulfide
gas) or reducing environment (alkaline-producing, bisulfide ion). If the
microorganisms responsible for sulfate reduction are still active, the pH
GEOCHEMISTRY 183
will remain between 6.5 and 8.0 because hydrogen sulfide gas is also toxic
to the sulfate-reducing bacteria. Methane generating bacteria cannot com-
pete with sulfate-reducing bacteria but are present in the most reducing
environments (Eh typically less than-300m V), usually beneath the sulfate
reducing bacteria.
An awareness of the fact that hydrogeochemical processes occurring in
ASR wells will tend to alter the pH and oxidation-reduction potential is
fundamental to the successful design and operation of ASR systems,
particularly in geochemically sensitive aquifers. Both changes can signifi-
cantly affect solubility and precipitation reactions, which have a direct
effect upon plugging rates, recovered water quality, and possible pre- or
post-treatment requirements.
Cation Exchange
Oxidation
Dissolution Reduction
Leaching
\ Decrease
Recharge
G)
Jl
Calcium 0
c
Eh z
Nitrate ~ NaCI 0
Increase Decrease ~
Total Dissolved Solids Sulfate m
............__ Increase Jl
Sodium ------. Chloride Jl
Bicarbonate Increase m
Sulfate Iron Sulfide ()
Manganese Iron I
)>
H2S. Sulfide CH4. C02 Jl
Bicarbonate G)
m
Fluoride
Phosphate
--- )>
z
pH 0
::E
m
r
Figure 5.5 Changes in groundwater chemistry with depth. r
CfJ
GEOCHEMISTRY 185
Eh-pH Diagrams
0.8
0.6
0.4
0.2
-
Ui"
0
z.. 0.0
.s:::.
!.U
-0.2
-0.4
-0.6
8 10 12 14
pH
>:mrmmn
.........
-0.20
·.·.·.·.·.·.·.·.·.
-0.40
Water Reduced
-0.60
Source: Hem. 1985.
pH
Figure 5.7 Equilibrium activity of dissolved iron as a function of Eh
and pH.
Unlike iron, the kinetics for manganese are extremely slow. Figure 5.8
shows an Eh-pH diagram for manganese for the same wells shown in
Figure 5.7 for iron. The Eh and pH relationships that appeared out of
equilibrium in the iron Eh-pH diagram indicate equilibrium with manga-
nese. This is a common relationship. Dissolved manganese from a filtered
sample is usually a more reliable indicator of Eh-pH conditions in the
GEOCHEMISTRY 189
Water Oxidized
Ponderosa
Wellfield
MnooH
Mn 30 4
MnC0 3
Mn(OH) 2
Dissolved
Manganese
Water
Reduced
If)
If)
~
Source: Hem, 7985. If)
If)
' If)
11)11)
0 2 4 6 8 10 12 14
pH
~
.c 0.2
w
-0.2
-0.4
-0.6
-0.8
0 2 4 6 8 10 12 14
pH
Once water samples for the potential ASR storage zone and for the
recharge water source have been obtained and analyzed, the next logical
step in the geochemical assessment is to use a geochemical simulation
model to evaluate in a preliminary manner whether mixing these two
waters in various proportions can potentially result in precipitation reac-
tions that may plug the aquifer. Such models are discussed subsequently
in Section 5.4, Geochemical Models.
When combined with local ASR experience at other operational sites,
the results of this assessment may indicate that geochemical issues are
unlikely. In this case it is appropriate to proceed directly to construction
1.2
N-0-H
25°C, 1 Bar
1.0
0.8
0.6
0.4
~
.t: 0.2
UJ
0.0
-0.2
-0.4
-0.6
pH
Figure 5.10 Nitrogen - Eh vs. pH.
192 GROUNDWATER RECHARGE AND WELLS
and testing of ASR facilities. In many cases, however, the results are not
conclusive, reflecting lack of data regarding aquifer mineralogy, and lack
of local ASR experience demonstrating that geochemical problems do not
occur. It is usually appropriate in such cases to obtain core samples of the
aquifer under consideration for ASR storage, in order to characterize the
physical, chemical, and hydraulic characteristics of the environment in
which the recharge water will be stored.
Continuous wire line cores of unconsolidated sediments or consolidated
rock, through the full thickness of the storage zone, provide the opportu-
nity for detailed lithologic analysis and comparison to geophysical logs
obtained after completion of the initial test hole at any proposed ASR site.
Core intervals may then be selected for analysis, usually one each in the
overlying and underlying confining layers, plus at least three in the storage
zone, depending upon aquifer heterogeneity as judged from the lithologic
analysis and geophysical logs. In the event of uncertainty, it is usually wiser
to select a greater number of cores for detailed analysis, ensuring that they
represent a reasonable balance of different lithologies in the storage zone.
Selected core intervals, usually about 30 em (12 inches) in length, are
then prepared for shipping to a core laboratory. This may involve rapid
freezing in dry ice, encasing in wax, or other approaches. If coring occurs
over an extended period, cores will need to be wrapped in clear plastic
sheeting; marked (top and bottom); labeled and stored appropriately, such
as in core boxes, in sealed PVC tubes, or frozen. In this case, core selection
may entail cutting previously frozen cores prior to shipping. Most of the
operational ASR sites in the U.S. for which core testing has been per-
formed, utilized frozen cores.
Cores obtained from the test well but not sent to the core laboratory
should be stored for the duration of the ASR test program and, preferably,
for a longer period that would include the first three years of operation. If
subtle geochemical shifts or changes in recovered water quality occur later
during ASR operation, supplemental core tests can be performed to verify
reactions and evaluate well response to potential remedial measures.
Core laboratories are typically capable of running a wide variety of
analyses required for petroleum industry operations. Analyses that have
proven useful for ASR projects are discussed in the remainder of this
section, including the following:
• physical characteristics
• colored pictures of cores
• mineralogy (X-ray diffraction, XRD)
• cation and base exchange capacity
• scanning electron microscopy (SEM)
• thin section petrography
GEOCHEMISTRY 193
Physical Characteristics
Mineralogy
Sample
Identification Plagioclase
Number Quartz Feldspar Calcite Dolomite Siderite Pyrite Kaolinite Illite/Mica Smectite Feldspar Feldspar K
CEC SurfaceArea
Constituent (meq/L) (m 2/g)
Zeolites 230-620
Organic matter 150-500 >1000
Mn0 2 (pH 8) 260
Soil organic matter >200
Vermiculite 100-150 750-800
Smectite 80-150 600-800
Montmorillonite 80-150 600-800
Saponite 80-120 700-800
Soil micas 20-60
Soil clay 5-60 150-200
Halloysite (4HP) 40-50 21-43
Chlorite 4-47 25-40
Illite 10-40 100-200
Soil clay loam 4-32 150-200
Palygorskite 20-30
Soil silt loam 9-27 50-150
Various soils 4-25
Soil loam 8-22 50-100
Glauconite 10-20
Soil sandy loam 2-18 10-40
Kaolinite 3-15 10-20
Halloysite (2H 2 0) 5-10
Soil sand 2-7
Oxides and hydroxides 2-6 14-90
Pyrophyllite 4
Basalt 0.5-3
Quartz, feldspars 1-2 1-3
A. 50 X
B. 200X I 50!-' I
c. 500X
D. 2000X 5 !..I I
In Figure 5.12 [6] the units of TSS are expressed in terms of milligrams
per liter TSS per square centimeter area of sand face. As little as 0.1 mg/
L TSS per square centimeter of sand face within the ASR well can cause
significant reduction in recharge rates at the well.
It is frequently useful to obtain a 0.45-f..Lm filtered sample during the
backflushing operation at an ASR well to determine the mineralogy of the
TSS. This, in tum, identifies sources of the solids clogging the well.
Freshwater diatoms were the most abundant at one ASR site in Seattle,
retained in the water from a lake which was used for recharge. Another
ASR site at Swimming River, NJ, had an aluminum-sulfate composition
resulting from operation of the water treatment plant. Others involved
reversing flow in old, unflushed municipal water lines. Each of these
problems can be addressed satisfactorily once the origin is known.
Biofouling
Bacteria are present in aquifers to depths of at least 500 m (1,500 ft) and
probably deeper. Many years ago it was believed that bacteria were only
present in shallow aquifers as a result of septic systems. Bacteria were
sampled by collecting filtered water samples. We now know that most of
the bacteria in the groundwater are attached to aquifer particle surfaces
and, therefore, are poorly sampled using groundwater as the sampling
media. Cores are required. Even with cores, little is known about the role
these bacteria perform in the groundwater system. Extrapolation from the
reasonably well-known role in the soils indicates that the bacteria may be
either directly or indirectly responsible for most of the changes in ground-
water chemistry in ASR wells.
As shown by Dragun [7], bacteria are typically present in surface soils
at populations of 0.1 to 1 billion cells per gram of soil and may be present
at populations as high as 10 billion cells/gram. In groundwater, bacteria are
typically present at concentrations of 100 to 200,000 cells/ml, and possibly
as high as 1 million cells/mi. Biofilms form around various sizes of sand
grains and glass spheres between 0.12 and 1 mm in diameter, reaching
equilibrium thickness of about 10 to 60 Jlm between 8 and 10 days. Several
studies have addressed the pH sensitivity of bacterial species and activity.
Figure 5.13 [8] shows the major parameters and their correlations that
affect the growth of microorganisms at or near the wellbore, another
potential fatal flaw for ASR projects. A disinfectant residual should be
maintained in the recharge water and also in the wellbore during storage
periods, to control and preferably eliminate the growth of microbiota at
and near the wellbore. Letting a well sit idle when not in use can result in
the growth of microbiota at and near the wellbore as a result of no
Gl
m
0
0
I
m
:s::
100~~~-=----.-----------.-----------.---------~ en
--i
:0
Source: Bichora, 7986. -<
0
Cll
g> 40 1\ '/ '. 1t'o ~ '=~~- ..... ~ 1\ ---1
'E
Cll
~
Cll
Q. 20 ~ ........................................... l
0~----------~------------~------------~----------~
0.0 0.1 0.2 0.3 0.4
Weight of Suspended Solids Added/Area of Sand Face (grams/cm 2 )
Figure 5.12 Change in flowrate during recharge with different concentrations of suspended
solids.
N
~
202 GROUNDWATER RECHARGE AND WELLS
Particulate Volume
Dissolved Oxygen pH
water in the well, resulting in reducing conditions and a very high dis-
solved carbon concentration from the degradation of the oil. These condi-
tions can result in an excellent environment for sulfate-reducing bacteria,
particularly if the groundwater is also under reducing conditions, or if
there is little movement between aquifers with different heads through the
wellbore. Sulfate-reducing bacteria can effectively clog a well screen.
Sulfate-reducing bacteria are only one of hundreds of different types of
bacteria that can cause biofouling. Most of the bacteria can only be
described by the methods used to determine their numbers. Their func-
tions, roles, and methods to control their numbers are virtually unknown.
Adsorption
ion Exchange
Carbon Dioxide
~6 C02+ 5 H20 ~ C3Hl1Ds.O,
C02
~ C02(nq) ~ H2C03 ~ HC03 ~ COj'
+CAL 2
'
Oxidation
Dissolution
TABLE 5.7
MINERAL SOLUBILITIES AND CONTROLLING MECHANISMS
100
Explanation
80 - - - 50"C
.z- - - 25"C
.,...
·:;;
- - - - O"C
<(
60
:s
~ H2C03aq
0
E 40
G)
~
G)
a..
20
Source: Hem, 1985.
5.0 6.0
pH
6.5
6.0
Note Silt to sand
particle sizes for
dolomitic limestone.
::r:
c.
5.5
5.0
4-8 Mesh
6 12 18
Months
Kinetics
Once pertinent data are available regarding water chemistry for both the
recharge water and the native water in the proposed storage zone, a
preliminary geochemical screening should be performed to identify poten-
tial fatal flaws for the ASR test program. Several relatively user-friendly
computer simulation models are available for this purpose, each of which
can be used to evaluate different proportions of recharge and native water
in order to determine whether precipitation or solution reactions have a
tendency to occur. Among these models are W ATEQ, PHREEQE, and
MINTEQ, as discussed briefly below.
Model results are no better than the available data. The poorest data are
usually critical. For the equilibrium models mentioned above, simplifying
assumptions include no biologic activity and no kinetics. Equilibrium
constants are limited and of variable quality. The models should be used
GEOCHEMISTRY 209
WATEQ
This name stems from the words "WATer EQuilibrium." The model
was developed by the U.S. Geological Survey and is the most user-friendly
of the equilibrium models. It has been thoroughly documented and evalu-
210 GROUNDWATER RECHARGE AND WELLS
PHREEQE
The name stems from "pH REdox EQuilibrium Equations." This model
was also developed by the U.S. Geological Survey and is fairly user
friendly, although most versions are user-modified. It is well documented
and evaluated. It includes aqueous speciation, solubility equilibrium and
also the reaction path, including the dissolution and precipitation of com-
mon minerals. The basic model includes 19 elements, 120 aqueous spe-
cies, rudimentary organics, 3 gases (carbon dioxide, oxygen and methane),
3 redox elements, ion exchange, and 21 minerals and compounds. It is
based upon temperatures between 0 and 100°C and 1 atmosphere pressure
[11, 12].
MINTEQ
EQ3NR/EQ6
Column Testing
These tests have been performed at three ASR sites: Myrtle Beach,
Chesapeake, and Swimming River. For each site, several frozen core
sections with diameters ranging from 2.5 to 4 inches (60 to 100 mm) and
lengths in the range of 12 to 24 inches (300 to 600 mm) were wrapped and
sealed with adhesive in clear plastic sheeting, with wire mesh on each end.
Steel end caps were machined to include tubing for movement of water
across the entire end face of the core, and also to include axial threaded
rods to restrain each end of the core. The wrapped core, screens, and end
caps were then supported on a small tray constructed of circular lexan pipe
and inserted into a 200 mm (8-inch) lexan cylinder with a length of 30 inches
(830 mm). Cylinder end plates, seals, and four tie rods were machined so as
to provide a pressure vessel capable of being pressurized to at least 30 psi.
The wrapped core was centralized along the axis of the pressure vessel, and
thrust restraint was then provided by the axial threaded rods.
Three cylinders were constructed of 8-inch (200 mm) PVC pipe and end
caps, each with a capacity of about 15 G (57 L). One was provided with
a glycerine-water mixture. The second was provided with recharge water,
and the third with native water from the ASR storage zone. A nitrogen
cylinder at 2000 psi with a regulator was also provided, along with
accurate flowmeters, pressure gauges, and sampling ports for water at each
end of the cylinder.
212 GROUNDWATER RECHARGE AND WELLS
Once the pressure vessel was sealed with the core inside, it was filled
with the glycerine-water mixture and pressurized, allowing sufficient time
for the core to thaw and reach thermal equilibrium. Recharge water was
then passed through the core at an inlet pressure below the confining
pressure surrounding the core. Samples were collected of the recharge
water and also the water leaving the core. The sampling frequency was
established so that a relationship could be determined between water
quality and the number of estimated pore volumes. Pressures and flow
rates were monitored, and observations were made regarding the appear-
ance of the product water, any apparent leakage through the core wrap,
temperature, and cumulative volume.
Some of the cores required a few hours to simulate a series of ASR
cycles, in each of which recharge water and native water were passed
through the core in opposite directions. Other cores required several days
to complete one simulated cycle. A few cores indicated leakage through
the end caps or other logistic problems, and had to be discarded. In general,
the equipment worked well but was expensive to construct to the required
tolerances and also expensive to operate due to the long duration of the
core testing and the skill level required of the operators.
Laboratory water quality data was then combined with flow and pres-
sure measurements and other observations to calculate trends in hydraulic
conductivity of the core material, and water quality of the flow leaving the
cores. pH adjustment of the recharge flows was tested on several cores, to
estimate the aquifer response to pretreatment measures under consider-
ation. Ion exchange, oxidation, precipitation, dissolution, and possible
bacterial reactions were evident in the quality of water leaving the cores.
Despite the expense and complexity of this effort, the results proved
useful in evaluating potential geochemical reactions in subsequent ASR
well construction and testing. Water quality data collected from ASR wells
and nearby observation wells tended to match results predicted from the
column tests. The column tests were less useful in predicting long-term
well hydraulic response to ASR operations since it was concluded that
bacterial activity, particle rearrangement or some other comparable mecha-
nism adversely affected hydraulic conductivity following a few days of
testing for almost all of the cores. The equipment may be useful for
evaluation of core and aquifer response to pre-flushing operations de-
signed to control potential clay dispersion in brackish aquifers; however,
this has not been investigated to date.
Figure 5.17 shows the column testing apparatus while Table 5.8 shows
typical results for a core obtained from an observation well at Swimming
River. The core was one of several subjected to column testing. It con-
tained a significant concentration of siderite (ferrous carbonate) and was
GEOCHEMISTRY 213
Note: Effluent samples collected as composites of the 2nd and 3rd pore volume (A), and sixth and seventh pore volume (B). Results: calcium declined from 22 mg/L in the
recharge water to between 3 and 5 mg/L in the effluent, iron was about 3 to 4 mg/L in the effluent, sodium increased from 36 mg/L in the recharge water to 87 mg/L
in the effluent after three pore volumes and 63 mg/L after seven volumes.
GEOCHEMISTRY 215
Batch Testing
This testing approach has been conducted for the Chesapeake ASR site.
Compared to column testing, this approach is relatively inexpensive.
Samples of core material are thoroughly mixed and are then subjected to
a progressive series of leaching tests at steadily increasing pH values to
determine dissolution, ion exchange, and other reactions occurring during
each step in the series. Results provide considerable insight into the
expected water quality and geochemical response to ASR operations.
However, they are not useful for directly evaluating aquifer hydraulic
response, except as may be inferred from the geochemical potential for
plugging or dissolution.
For the Chesapeake ASR site, the issue was manganese concentrations
in the recovered water which exceeded drinking water standards. No
manganese problems had been detected during previous ASR cycle testing
at pH values ranging from 7.4 to 9.4. However, subsequent ASR opera-
tions had been conducted at a lower pH (5 .9 to 7.4) relative to cycle testing,
causing manganese to become solubilized. Manganese concentrations in
the recovered water ranged from 0.25 to 1.28 mg/L.
The following sequence of batch-testing operations was conducted on
core samples selected from a repository created during the coring to be
representative of the storage zone, based upon geophysical logs, drillers
logs, and core laboratory data. During the first test, core samples from four
selected depths were sequentially analyzed to determine the manganese
concentrations associated with leaching, first in water at low pH and then
in ammonium acetate. From this test, the core with the greatest potential
for manganese release was identified. Analysis showed it contained man-
ganese oxide, organic manganese and 137 mg/kg of manganese associated
with iron oxyhydroxides. Another sample from this core was then split and
each portion tested, one with pH 6.5 water and one with pH 8.0 recharge
water for a period of 33 days, to determine the released manganese
concentrations.
Based upon batch testing results, pH adjustment of the recharge water
was implemented to control manganese reactions in the aquifer. An opti-
216 GROUNDWATER RECHARGE AND WELLS
6.1 ECONOMICS
217
N
......
CD
1. Unit costs for ASR facilities generally range from about $50,000 to $160,000/
megaliters/day ($200,000 to $600,000/MG/day) of recovery capacity, with
an overall average of about $100,000/megaliters/day ($400,000/MG/day).
Higher unit costs are typically associated with the first new ASR well at any
site, sites requiring extensive piping to tie them in to the existing water
system, and sites with low recovery capacity per well. Lower unit costs are
associated with retrofitting existing wells at sites close to existing piping
facilities, higher yield wells, and also with multiple-well ASR expansion
projects.
2. Unit costs for the second and subsequent ASR wells at any site are typically
lower than for the first well, reflecting generally reduced efforts to obtain
regulatory approval. The first ASR well incurs additional cost in order to
demonstrate ASR feasibility. The reduction in unit cost is typically in the
range of $26,000 to $53000/megaliters/day ($100,000 to $200,000/MG/
day).
3. When comparing capital cost per unit of new capacity, ASR typically is less
than half the cost of other water supply and treatment alternatives. In some
220 GROUNDWATER RECHARGE AND WELLS
cases, the cost savings approach 90%. This savings reflects the efficient use
of major facilities such as pipelines, pumping stations, and treatment plants,
and the relatively low cost of using underground storage capacity when
compared with a similar storage volume provided in surface reservoirs.
4. Annual operating cost ranges from about $1,600 to $10,600/megaliters/day
($6,000 to $40,000/MG/day) of recovery capacity, although data availabil-
ity is sparse. This includes the marginal cost for power and chemicals during
recharge and recovery, plus an allowance for operation and maintenance.
One of the three sites is somewhat atypical in that treated drinking water is
stored in a seawater aquifer at Marathon, FL. Consequently typical annual
operation costs based upon this data are probably closer to about $15,000/
MG/day recovery capacity.
Wyoming, Ml 9 31
Peace River, FL 46 108
Manatee County, FL 2 38
Florida Keys, FL 3 38
Kerrville, TX 3 30
The savings in capital costs provided through ASR create the opportu-
nity for a corresponding reduction in water costs and associated rates. A
50% reduction in the capital cost of facilities expansion can lead to a
corresponding reduction in water rate increases required to finance the
expansion. In some cases, the need for an expansion program and associ-
ated rate increase can be deferred for several years, reduced or eliminated
through more efficient use of existing facilities.
Where rate increases are required for systems utilizing ASR wells, it is
pertinent to consider certain fiscal questions that are unique to an ASR
mode of operation. Two key questions are as follows:
222 GROUNDWATER RECHARGE AND WELLS
• If a water utility system supplies several water users, meeting their demands
with a system that incorporates ASR, who should pay for water stored but
not recovered within the same fiscal year and when should they pay?
• If there are multiple users, some of whom wish to recover water to meet
short-term seasonal needs, and others who have longer-term needs with
varying storage periods, or emergency needs, how should the associated
costs be apportioned among the users?
from local aquifers at much lower costs. To ensure the long-term viability
of such a program, it is important that the seasonal cost incentive matches,
or does not exceed, the avoided costs for regional peaking facilities.
Where the water utility provides both the water and the ASR storage,
the concept of a seasonal rate incentive is internalized within the utility and
its rate base. Awareness of the approximate marginal cost for water stored
and recovered, and the more costly alternatives available for meeting peak
demands, can probably guide ratemaking in an appropriate direction.
For some utilities, ASR storage within the service area provides the
opportunity for negotiation of favorable terms in wholesale purchase
agreements for imported water. Water can be purchased during periods of
low demand and/or high supply, thereby achieving low marginal costs.
Transmission facilities may possibly be downsized, reflecting planned
conveyance of water during off-peak months at lower flow rates, so that
peak demands can be met locally from ASR storage.
A significant challenge in establishing rates with ASR is to overcome
the traditional utility rate-making process thinking in which a major water
user perceives that he is buying "system capacity" on a "take-or-pay"
basis. With ASR, it is not as simple as paying some money and having
access, whenever needed, to a fixed amount of facilities capacity. Instead,
the water user is buying a commitment to supply water at some time in the
future, whether months or years away. This requires reasonable planning
and forecasting of water needs so that adequate facilities can be built and
operated in time to store sufficient water so that it will be available when
needed. The payback for this planning is the reduction in water costs, since
facilities for treatment, storage, and conveyance can be substantially
downsized if ASR wells are used to help meet future peak demands. This is
where the 50% or greater reduction in capital cost amortization is achieved.
Conversely, if a water user insists on a conventional type of rate agreement,
the facilities will probably have to be built up front to meet that full demand
without ASR storage and the cost will therefore be much higher.
The role of ASR in the overall water supply plan is a concept that is not
always easy to understand. When elected officials are faced with the need
to establish rates and make commitments based upon water demand pro-
jections, it is frequently easier to defer construction of ASR and other
facilities and thereby hold down short-term costs. This can reduce that
water storage volume that can be achieved. It can also limit the opportunity
to reduce the size of the associated treatment facilities, thereby increasing
long-term costs to the water users.
The impact of ASR upon water rates is perhaps less significant than
other factors potentially affecting water rates. By making more efficient
use of existing facilities and aquifer storage, ASR helps to hold down
SELECTED ASR NON-TECHNICAL ISSUES 225
overall costs, but may not substantially affect the rate structure or its
underlying philosophy. Where ASR is implemented to augment water
supplies to meet increasing projected demands, the same demands can
perhaps be met through a sharply inclined rate structure over and above a
base monthly volume per household. The resulting water conservation
effect can also extend the service life of existing facilities.
The principal frontier for ASR activity in the next few years is probably
associated with legal and regulatory issues at the federal and state levels.
How these issues are understood, framed, and resolved will probably
determine the extent to which ASR technology can ultimately benefit
water users throughout the country.
As an aid to those who are considering how to regulate ASR activities,
and the appropriate water quality standards to follow, Appendix A con-
tains the current (1993) standards adopted by the U.S. Environmental
Protection Agency, the European Community, and the World Health Or-
ganization to govern drinking water quality.
The draft EPA groundwater rule was announced in July 1992. Formal
proposal of the rule is currently scheduled for August 1994, with its
promulgation 24 months later. EPA proposes that all public water systems
using groundwater disinfect the source water from its wells unless (1)
SELECTED ASR NON-TECHNICAL ISSUES 227
"natural disinfection" requirements are met, or (2) the system qualifies for
a variance. The two exceptions are intended to include only those systems
where wells are not vulnerable to viral contamination. All sources would
have to provide a disinfectant residual during distribution.
The regulatory position related to recovery of treated drinking water
from ASR systems has not been determined. Most ASR sites are in deep,
confined aquifers and are not likely to be affected by overlying land-use
activities. Consequently, it should be possible to demonstrate the absence
of viruses and pathogenic bacteria in ASR recovered water, thereby meet-
ing the "natural disinfection" requirements when they are implemented.
It is quite clear from this wording that the intent of the act is to ensure
that no injection practice should occur if it would jeopardize the ability of
a water treatment plant treating water from the same aquifer to remove any
contaminants prior to distributing treated water for public consumption.
In the process of developing regulations to implement the act, the U.S.
Environmental Protection Agency (EPA) interpreted this language as
meaning that water should meet primary drinking water standards prior to
injection into the well. These standards are generally becoming more
restrictive. The UIC regulations provided for individual states to accept
responsibility for implementation of the UIC program so long as state
standards are at least as strong as the federal standards. As a result,
approximately 39 states have developed their own UIC regulations which
are patterned after the federal regulations, and have therefore accepted
responsibility for implementation of the act.
The federal UIC program includes classification of injection wells into
five categories: Classes I through V. ASR wells are generally considered
as Class V wells, which include a very wide variety of injection practices
other than ASR. Some of these practices represent a significant threat to
groundwater quality.
The UIC program also includes an aquifer exemption process that
provides for those situations where the quality of the injected water does
not meet primary drinking water standards. The process depends upon the
quality of water in the receiving aquifer. If the TDS (total dissolved solids)
concentration exceeds 3000 mg/L, a minor aquifer exemption is required
that can be handled at the state and regional level (one often EPA regions
around the country). However, if the TDS concentration in the aquifer is
less than 3000 mg/L, this is considered a major aquifer exemption requir-
ing action at the EPA headquarters level. To date, EPA has approved very
few minor aquifer exemptions and no major exemptions for Class V wells.
Minor exemptions have been generally issued for regional areas rather
than for individual wells. Exemption criteria are listed in Section 146.4 of
the 40 Code of Federal Regulations (40 CFR).
The issuance of such exemptions is understandably sensitive from a
regulatory point of view. Once contaminated, potable aquifers require
many years to restore their water quality, if full restoration of quality is
achievable. Consequently, regulatory positions have generally been con-
servative and consistent in requiring treatment to potable standards prior
to injection into Class V wells and potable aquifers.
Currently, ASR storage of treated drinking water in fresh or brackish
aquifers is generally considered to be an acceptable practice, as evidenced
by the growing number of operational ASR systems in the U.S. Ten states
SELECTED ASR NON-TECHNICAL ISSUES 229
provide for ASR implementation where water quality meets these stan-
dards with the possible exception of certain constituents that do not
threaten public health and the environment; are readily treated either in
the aquifer or in public water treatment facilities; and meet the original
intent of the 197 4 AcL In this way, EPA can approve state-level ASR
programs; continue with the UIC program implementation for non-ASR
systems; and help to attain the broad national, state, and local environ-
mental benefits associated with widespread ASR implementation. To
date, such a model legal and regulatory process does not exist. However,
it is the subject of considerable activity in several states and may emerge
in a form suitable for consideration by other states and countries in the near
future.
Some of the constituents that may not meet drinking water standards but
may be quite suitable for ASR recharge purposes in some situations
include the following:
• colifonns
• turbidity
• color
• sodium
• chloride
• total dissolved solids
• iron
• nitrate
• corrosivity
It is important that any model code that provides for easier implementa-
tion of ASR should not be interpreted as an opportunity to authorize water
management practices that, in fact, contaminate the environment. For ex-
ample, poor quality urban runoff may contain many constituents that violate
drinking water standards, such as metals, oil and grease, and other param-
eters that may be carcinogenic, mutagenic, teratogenic, or otherwise delete-
rious. ASR storage of this water in a freshwater aquifer would be unaccept-
able. Seasonal storage of urban runoff in a brackish aquifer following
pretreatment, such as detention/retention, may be a beneficial water manage-
ment practice in some situations. The legal code and associated regulations
adopted for any particular state or region may follow a model code or may
follow the example established in another state; however, it will undoubt-
edly reflect local priorities for water use and water quality protection.
Suggested elements of such an ASR Model Code include the following:
western states, many of which have fewer water resources, a prior appro-
priation water law doctrine is followed.
For ASR projects in both eastern and western states, the ownership of
the stored water is an important issue. In general, experience is supporting
the position that, if the water user has the right to the water prior to ASR
storage, he also has the right to recover that water. In other words, the state
provides some protection of his right to store the water. However, in some
states groundwater law is not adequately defined. In these states it is
theoretically possible for another water user to construct a well adjacent to
the ASR facility and pump out the stored water. Alternatives available to
a water user to protect his rights to the stored ASR water include location
of the ASR facility a sufficient distance from property lines that the risk
is minimized; municipal zoning or land-use control in the vicinity of the
ASR site; local municipal ordinance, or changes in the state law to
provide for ASR storage. Fortunately, it is reasonably common for the
radius of the storage bubble around the ASR well to be quite small,
frequently facilitating judicious siting of ASR facilities to minimize or
eliminate this problem.
As a part of the regulations, states may impose constraints upon re-
charge and recovery operations that go beyond water ownership. These
may include consideration of impacts upon other water users who may be
in existence at the time the ASR system comes on line. Recharge and
recovery rates, or selection of storage zones, may be regulated so that
unacceptable adverse effects upon existing legal users of the water re-
source do not occur, or are mitigated.
There is a need to integrate groundwater, surface water, and ASR
permitting so that the situation is avoided in which a water user obtains the
right to divert, treat, and store surface water underground, but is unable to
recover the stored water at desired recovery rates due to groundwater
permitting restrictions. It is usually desirable to recover all of the stored
water. In some cases, such as in areas of depleted groundwater resources,
it may be desirable to leave a small percentage of the stored water in the
ground to slowly replenish the resource until such time as target water
levels are reached.
Initial ASR permits are sometimes issued providing only for seasonal
storage and recovery, so that annual volume recovered cannot exceed the
234 GROUNDWATER RECHARGE AND WELLS
volume stored in the same year. However, greater benefit can be achieved
by writing the permit to provide for carryover storage from one year to
another. In this way, excess water available during wet years can be carried
over to meet needs during drought years. Similarly, water stored during
early years following a water treatment plant expansion can be carried over
to meet increasing peak demands in later years when treatment plant
capacity may be insufficient. If long-term storage is feasible and permit-
ted, it can defer and downsize the next facilities expansion phase, achiev-
ing considerable savings. A preferred approach is to regulate according to
long-term cumulative storage rather than seasonal or annual variability.
A related issue is the desirability of providing for seasonal allocation of
recharge water. It is quite common for recharge water to be available
according to an annual volume allocation, based upon safe sustained yield
of the water source during the dry season. An ASR allocation process
could be seasonal, making more efficient use of water available during wet
months.
Recovery Percentage
ASR wells are used for both injection and recovery. In addition to
enabling the well to be periodically backflushed, thereby maintaining its
injection capacity, this approach also facilitates hydraulic control of the
storage bubble around the well, thereby minimizing mixing with surround-
ing native groundwater. Furthermore, this approach is quite cost-effective
in that the same facilities are used for both recharge and recovery. Regu-
latory approval of ASR facilities and operations has been facilitated by
these inherent advantages.
However, some situations may arise where it is desired to recharge at
one location and recover at another location. One possible reason for this
is the need for blending two different water qualities to achieve a relatively
uniform product water quality. This can be achieved by designing the
location of injection and recovery wells and well-head facilities with
enough spacing and capacity as to achieve the desired blending ratio. A
more common reason is to use the aquifer as a means of conveyance and
also long-term storage. Water injected at point A is recovered from the
same aquifer at point B, even if the distance between the two points is such
that the travel time may be hundreds of years or more. The net volume of
water in the aquifer, and associated water levels, are maintained through
artificial recharge practices.
This is a sound water management approach; however, it is not ASR. Two
fundamental issues challenge the widespread application of this approach.
First, injection wells tend to plug unless they are periodically redeveloped.
The redevelopment frequency and associated cost may be such that an ASR
design for the injection well would provide better service since the pump in
the injection well would help to maintain injection capacity. Second, water
rights laws in some western states that follow the prior appropriation legal
doctrine can inhibit the ability of a water user to operate recharge facilities
in this manner. For example, such a practice is legal in California, Arizona,
Texas, and possibly other states but is not legal in Colorado.
ASR Education
A common theme among the various ASR legal and regulatory issues
and processes is a lack of understanding of ASR technology, applications,
and experience. In some cases, this extends to significant misconceptions
that must be overcome before permitting can be completed successfully.
This is not surprising in view of the recent development of this technology.
In 1983, only three ASR systems were operational, whereas in 1994, at the
time of writing, 20 such systems were operational and about 40 additional
systems were under development in 13 states. Successful completion of
ASR permitting requires an initial effort to educate regulatory personnel
regarding the technology and its operational experience at comparable
sites. This is best accomplished when the Phase 1 ASR feasibility report
is completed; this provides the opportunity to present report findings and
recommendations to regulatory staff, including well-thought-out plans for
field investigations to be completed under Phase 2.
addressed include ownership and control of the stored water, how far the
storage bubble extends from the well, how far the water level effects
extend from the well during recharge and recovery, how much of the
stored water will be recovered, how much the program will cost, and the
associated potential costs and savings for the ratepayers. Attendees at
these meetings are frequently well informed and can pose sophisticated
questions.
Where storage is planned in brackish or non-potable aquifers, it is also
necessary to discuss mixing between stored and native water, recovery
efficiency, and disposal of brackish water to the environment during
drilling and testing. Where sources other than drinking water are under
consideration, water quality impacts will have to be addressed. These
issues require great care and excellent graphics in order to avoid misunder-
standing.
Public support for ASR programs has generally been strong. Opposition
has generally been due to support for other water supply alternatives, such
as surface reservoirs and major transmission pipelines, which may have
considerable, long-standing momentum in the community. Sensitivity to
these underlying issues can frequently provide guidance as to how to
handle them.
It is quite common for serious consideration of ASR as a water manage-
ment alternative to begin after years of effort to construct a surface
reservoir or long transmission pipeline against increasingly strong envi-
ronmental and other opposition. Polarization of supporters and opponents
to the surface reservoir or pipeline project is frequently far advanced by the
time that ASR enters the discussion. It is usually unwise to tackle this issue
directly since surface reservoir or pipeline momentum is frequently strong
and supported by the same individuals who would ultimately need to
support a change in direction toward ASR. ASR is therefore in the position
of presenting a viable and cost-effective alternative to their preferred long-
standing project, at a time when this may be unwelcome for a variety of
reasons.
In such situations, the only viable recourse is usually to emphasize the
need for site-specific field investigations to confirm feasibility and eco-
nomics before any serious consideration of ASR to meet local needs. This
may enable the ASR program to move forward quietly, without reducing
momentum for the alternative project. In some cases, it is also appropriate
to point out the merits of combining both projects, assuming that the
surface reservoir or pipeline will definitely be built. In many cases an
excellent case can be made that the combination of ASR and surface
storage can work together to achieve benefits not attainable with either
system by itself.
SELECTED ASR NON-TECHNICAL ISSUES 239
Alternative
ASR
Applieations
"From water, God made every living thing:·
Surah 21, At Anbiya, The Koran
7.1 INTRODUCTION
With the growing success and acceptance of ASR for storage of drink-
ing water supplies, attention has begun to focus on the potential applica-
tion of this technology for storage of water from other sources, such as
untreated or partially treated surface water from rivers, canals, lakes, and
reservoirs; untreated groundwater; and reclaimed water from wastewater
treatment plants. Common to each of these other sources is that they are
generally of a quality that falls short of meeting drinking standards. This
presents a challenge, not only in technical terms but also for regulatory
reasons.
Due to the immense potential benefits associated with ASR storage of
water from these sources, it is anticipated that considerable effort will be
devoted in the future to the resolution of technical constraints and reevalu-
ation of the existing regulatory framework governing such applications.
Where such waters can be safely stored underground without significant
241
242 GROUNDWATER RECHARGE AND WELLS
risk to the environment, water quality or public health, these benefits can
be realized so long as the associated projects can be permitted with
reasonable effort and cost. On the other hand, there is no justification for
underground storage of waters of such low quality that wells may clog or
that aquifer contamination may occur. Much work remains to be done;
however, progress is underway in several states. In the remainder of this
chapter, the current status of ASR applications for non-potable water
sources is discussed.
Technical Considerations
Regulatory Considerations
Economics
connections and other facilities to convey the water from the wastewater
treatment plant to where it is needed, and to store the water at times when
it is not needed. Storage requirements depend upon the variability in
supply and demand for reclaimed water during the year, and also upon the
viability of discharging the water to waste during periods of heavy rainfall
or other times when it is not needed.
At some locations, ASR storage of reclaimed water in brackish aquifers
may be viable and cost-effective, storing when excess reclaimed water is
available and recovering when irrigation or other demands exceed avail-
able supplies. By providing inexpensive storage of reclaimed water, ASR
may help to expedite funding and implementation of expensive regional
plans to make better use of reclaimed water for beneficial purposes.
California has led the way in the development of draft state regulations
(July 1992 draft) governing the use of reclaimed water for recharge of
potable aquifers, whether through wells or surface recharge facilities.
These draft regulations have been developed over a period of several years
and reflect considerable effort by many experts. As a result, they provide
useful experience to guide those interested in developing comparable
regulations for other areas.
A principal component of the California draft regulations is to ensure
that water produced at nearby domestic production wells in the same
aquifer does not exceed specified percentages of reclaimed water. Re-
quired blending ratios at the nearby production well depend upon the
associated level of treatment for the reclaimed water. Under the draft
regulations, all recharge waters would have to undergo biological oxida-
tion and disinfection, with well injection also requiring filtration and
organics removal. Surface recharge would require filtration when the
water of reclaimed origin may exceed 20% of the total flow from any
domestic water supply well in the vicinity, and would require filtration
plus organics removal when this percentage exceeds 50%.
Oxidized wastewater must not exceed 20 mg/L total organic carbon
(TOC), 30 mg/L suspended solids and 30 mg/L biochemical oxygen
demand. Filtered wastewater is defined as not exceeding an average of 2
nephelometric turbidity units (NTU), and not exceeding 5 NTU more than
5% of the time during any 24-hour period. Disinfection requirements
depend upon the level of prior treatment. Organics removal depends upon
the method of recharge and also upon the reclaimed water contribution as
a percent of total flow from any domestic water supply well in the vicinity.
252 GROUNDWATER RECHARGE AND WELLS
For surface recharge, draft TOC limits vary from 20 to 6 mg/L, while for
well injection they vary from 5 to 2 mg/L. A minimum spacing of 2000 ft
between the injection well and the nearest domestic supply well is speci-
fied. Travel time between the two wells has to exceed 12 months.
The TOC requirements for reclaimed water that is injected directly into
an aquifer are designed to prevent more than 1 mg/L ofTOC from reaching
domestic supply wells. If the TOC in the reclaimed water is 5 mg/L, 20%
of the water pumped by a nearby domestic supply well can be reclaimed
water. If the TOC of the injected water is 2 mg/L, then the nearest domestic
supply well can get up to 50% of its water from the reclaimed water source.
The regulations assume no TOC reduction in the aquifer.
The cost of reducing TOC to meet well injection criteria can be
substantial, reflecting treatment by such organic removal processes as
granular activated carbon adsorption and membranes. The difference be-
tween organic removal requirements for surface recharge and well injec-
tion is primarily based upon the demonstrated efficiency of aerobic bio-
logical processes occurring in the vadose zone to remove such organics,
and the presumed absence of such processes in the saturated zone below
the water table.
As discussed previously in Section 4.6, Disinfection Byproduct Reduc-
tion, organics removal does occur in the saturated zone based upon recent
investigations. Disinfection byproducts are only a small component of the
entire range of organic compounds in the aquatic environment. However,
their significant reduction during aquifer storage under confined, saturated
conditions suggests the possibility that other organics may also be re-
moved in a similar manner, probably due to biological activity in the
aquifer and possibly to adsorption processes. It is too soon to draw conclu-
sions regarding the efficiency of a saturated aquifer system to reduce
concentrations of a broad range of organics, however, the suggested direc-
tion for further research is clear. Similarly, the fate of organic compounds
around an ASR well needs to be established. Breakdown into innocuous
compounds such as carbon dioxide, nitrogen, and water would undoubt-
edly be satisfactory; however, breakdown into other organic compounds
may be less desirable. Further understanding of organic removal processes
and the fate of these compounds in a saturated environment will provide
a basis for the design and operation of facilities to optimize organics
removal during aquifer storage. It will also provide an improved basis for
assessment of the risks and benefits associated with seasonal storage of
reclaimed water in brackish aquifers.
Recent studies in St. Petersburg, FL, where reclaimed water is injected
into a 213 to 338 m (700 to 1110 ft) deep saline injection zone, revealed
ALTERNATIVE ASR APPLICATIONS 253
that the hydrogeologic system within this and the overlying brackish
aquifers is very reactive from a geochemical and biochemical viewpoint
[1]. Results were similar to those for several case studies at other injection
sites in the U.S., as determined from a literature search. In particular,
metals, nutrients, and trace organics are removed by a variety of processes
in these confined aquifers, including precipitation, ion exchange and bac-
terial activity. Generation of methane and nitrogen gas is a common
metabolic byproduct of these processes. Particularly active areas in injec-
tion zones occur within the immediate vicinity of injection well boreholes
and within the injection front of the fluids that were injected.
from wet years to dry years would be allowable only to the extent that
monitoring shows that migration of the storage bubble away from the well
is not significant under the natural hydraulic gradient prevailing at the site.
6. Treatment of reclaimed water prior to injection would be as needed to meet
quality standards and minimize plugging of the well due to particulates or
other constituents. Treatment would generally need to include oxidation,
filtration, and disinfection. At most sites, aquifer hydraulic characteristics
will be such that high quality reclaimed water will be necessary in order to
minimize plugging and ensure satisfactory long-term well operation.
7. Concentrations of organic compounds in the reclaimed water would be
reduced during seasonal aquifer storage. Since native water in the aquifer
exceeds potable TDS standards, any recovery of residual stored water
blended with native aquifer water in a nearby domestic supply well would
probably require membrane treatment to meet potable standards. The time
period for movement between the injection well and the nearest domestic
supply well, combined with the lateral distance between these wells, would
constitute effective barriers to protect public health. In addition, membrane
treatment of the brackish water recovered from the domestic supply well
would provide a third organics removal barrier to enhance public health
protection. The regulatory process would define the extent of these barriers by
identifying lateral distances between the ASR well and other domestic wells
potentially using brackish water from the same zone for drinking purposes.
8. Nutrients in the reclaimed water may benefit irrigation water users; how-
ever, during ASR storage they would probably promote bacterial activity in
and around the well, potentially causing well clogging. Whether this can be
effectively controlled by maintaining a chlorine residual in the well, as with
potable water ASR, remains to be confirmed. Until such time as this question
is resolved, a suggested approach is to maintain a chlorine residual in the
reclaimed water being recharged, while that water is moving through the well
and gravel pack. During storage periods, trickle flow a small chlorinated
supply of water into the well at a rate sufficient to maintain this residual in the
well. The required flow rate and residual concentration can be estimated by
periodically pumping samples from the well following a typical cessation of
recharge, to determine the decay rate of the chlorine residual in the well. The
intent is to preclude bacterial activity in the immediate vicinity of the well,
where it would have the greatest potential for clogging.
9. As discussed in Section 4.6, Disinfection Byproduct Reduction, initial
formation and subsequent significant reduction of trihalomethanes and
haloacetic acids should occur in the aquifer if the water is stored for a period
of several weeks, particularly at sites where anaerobic conditions develop
during storage and where a carbon source is available to support bacterial
activity.
Future
Directions
"Would you tell me, please, which way I ought to walkjl-om here.?"
"That depends a good deal on where you want to get to," said the Cat.
"!don't much care where-" said Alice.
''Then it doesn't matter which way you walk," said the Cat.
Alice's Adventures in Wonderland, Lewis Carroll
It is interesting to look back and see the progress that has been made
during the past few years with the development and implementation of
ASR. From a concept not many years ago it has evolved into a proven, cost
effective water management tool today. The ability to use wells to store
and recover a vast amount of water underground at low cost is a significant
advancement in water management. A "sea change" has occurred in how
we manage water. The change is already apparent to some, and will
become apparent to others in future years as the locations and applications
of ASR technology become more diverse. It is pertinent, therefore, to
consider what lies ahead. What new developments may affect the locations
and methods of ASR implementation in the future? What should be the
directions for future research in order to expand the benefits associated
with ASR? These and other issues are discussed in Chapter 8.
257
258 GROUNDWATER RECHARGE AND WELLS
It seems reasonable to expect that the next few years will bring improve-
ment in understanding of water quality changes occurring underground
during ASR storage. It is becoming increasingly clear that bacterial,
geochemical, physical, and other processes occurring in both saturated and
unsaturated formations are effective at improving water quality. Better
understanding of these mechanisms can enable improved design and op-
eration of ASR systems to achieve specific water quality goals. Until better
understanding is achieved, however, it is appropriate to take a conservative
position with regard to planning and regulation of ASR systems to store
water not meeting potable standards. Once contaminated, aquifers may
require a long time to restore water quality.
As an example of water quality improvement, organics such as disin-
fection byproducts (DBPs) have been shown to decline during ASR
storage as shown in Section 4.6, Disinfection Byproduct Reduction. The
mechanisms governing this have not been confirmed, although bacterial
activity is believed to be the driving force. Dilution and adsorption also
may be important mechanisms at some sites. Ongoing laboratory inves-
tigations are directed at improving the level of understanding of the
primary mechanisms, thereby providing a stronger basis for design and
operation of ASR systems to maximize DBP reduction where this is a
primary goal. Consideration should be given to the following future
endeavors to advance understanding of ASR organics reduction mecha-
nisms and applications:
• At new ASR sites storing water from non-potable sources, and in laboratory
column and/or batch testing investigations, gather data regarding reduction
under saturated conditions of other organics compounds besides DBPs.
• Develop a model code for regulation of potable water ASR systems. The
code would incorporate various key elements essential to ASR success,
while providing the flexibility to fit in with different legal frameworks for
water regulation in each state.
• Devise a regulatory framework for ASR storage of non-potable, high quality
waters in brackish aquifers that will work in Florida and California. These
two states will provide a pattern for other states to consider as they subse-
FUTURE DIRECTIONS 261
quently tackle the same regulatory issues. This would include concepts
presented in Section 5.3, Legal and Regulatory Issues, and Section 7.3,
Reclaimed Water.
• Seek the support of state environmental groups and major water users from
agriculture, industry and public water suppliers for the proposed regulatory
framework.
• Work with state regulatory agencies and elected representatives to imple-
ment rule changes as appropriate to establish the desired regulatory frame-
work at the state level.
• Seek Congressional action directing the Environmental Protection Agency
to modify existing Underground Injection Control regulations to provide for
ASR storage of high quality, but non-potable water in brackish aquifers,
developing a regulatory process that more closely reflects the benefits that
may be achieved without adverse environmental and water quality effects.
• Work with the Environmental Protection Agency to develop an appropriate
ASR regulatory process. This will require careful negotiations to establish
consistency with the intent of the 1974 Safe Drinking Water Act, even
though the Underground Injection Control regulations established pursuant
to the Act remain in effect. This will probably require development of
additional federal regulations pertaining to permitting of ASR facilities.
• Select sites for demonstration ASR projects storing surface water, groundwa-
ter, and reclaimed water, using the permitting process for these sites as the
vehicle for bringing about the necessary regulatory changes discussed above.
This process may require several years, with uncertain outcome. How-
ever, the competition for available water supplies will continue to build,
increasing the likelihood that changes will be necessary in the way we
manage water. These will include both technical and regulatory changes,
Driving Forces
• very low cost relative to other water supply and treatment alternatives
• relatively simple technology to design, construct and operate
262 GROUNDWATER RECHARGE AND WELLS
Constraints
Opportunities
Seleeted
Case
Studies
9.1 PEACE RIVER, FLORIDA*
* Peace River/Manasota Regional Water Supply Authority, 145 l Dam Road, Bradenton, Florida
34202
265
266 GROUNDWATER RECHARGE AND WELLS
stone artesian aquifers that contain brackish water. Well T -1 was con-
structed in the Tampa zone, which occurs at a depth of about 122 to 152
m (400 to 500ft), while the remaining five wells (Sl, SlA, S4, S5, S6)
were constructed in the Suwannee zone, which occurs at a depth of about
174 to 274m (570 to 900ft). Water is stored in the ASR wells during low
demand months and is recovered as needed to meet system peak and long-
term demands. Figure 4.18 shows the layout of the Peace River facilities.
Operation of the first two ASR wells began in 1985. Three wells were
added and an observation well converted to ASR use in 1988. Three
additional ASR wells should become operational in 1994, increasing
system recovery capacity to between 26 and 30 megaliters/day (7 and 8
MG/day). To meet seasonal and long-term variations in water supply and
demand, the target storage volume for each well is 1.33 Mm3 (350 MG) per
MG/day recovery capacity. As of the end of 1993, this target volume has
almost been reached since the Authority has 5.7 Bm3 (1.5 BG) of treated
drinking water stored in the ASR wells.
A third zone in the Avon Park aquifer is located at a depth of about 396
to 427 m (1300 to 1400 ft). This zone is being tested to determine its
feasibility for storage of untreated or treated water. If feasibility is con-
firmed, all three zones would be utilized to store water beneath the treat-
ment plant, thereby "stacking" the stored water vertically and concentrat-
ing piping and wellfield operations in a small area. This is quite cost-
effective. Figure 2.7 shows the hydrogeologic cross-section at this site.
Estimated hydraulic characteristics of the various aquifers and confin-
ing layers have been estimated from several pumping tests. For the Tampa
zone, transmissivity is about 455 m2/day (4900 ft 2/day); storativity is
0.0004, and leakance is 0.0001/day. Porosity is estimated at 15%. Static
water level is about 22 ft above the measuring point, which is at an
elevation of 27 ft above mean sea level. The average regional gradient of
the potentiometric surface in this area is about 0.0002 to the WNW;
however, production wells in the vicinity may affect the local gradient
around the single ASR well in the Tampa zone. This well is open hole in
limestone from 380 to 480 ft in depth. The lithology for Well T -1 is
generally limestone, poorly to well consolidated, light grey to cream
colored, soft to hard, and fossiliferous.
For the Suwannee zone, transmissivity is about 560 m2/day (6000 ft2/
day); storativity is about 0.0001, and leakance is 0.0085/day. Most of the
leakance is believed to occur through the underlying confining layer
separating this zone from the Avon Park formation. Static water level is
about the same as for the Tampa zone, and the regional gradient is about
0.0003 to the WNW. All but one of the existing operational ASR wells are
SELECTED CASE STUDIES 267
TABLE 9.1
RECHARGE AND NATIVE WATER QUALITY: PEACE RIVER, FLORIDA
Figure 9.1 ASR Well S-6, monthly recharge and recovery volumes and TDS concentrations,
Peace River, Florida. II)
a>
l!ll
270 GROUNDWATER RECHARGE AND WELLS
aereal extent of the wellfield. Each well site would include a pair of wells,
one in each zone.
* Department of Utilities/Public Works, City of Cocoa, 600 School Street, Cocoa, Florida 32922
SELECTED CASE STUDIES 211
Scale in Feet
liiJ"'iiil I
0 1500 3000
.,1::-"'c
.c 5
..
.n ,.,., $:Gi ~
UJ
:J m
t
0.c:: -~r::
~ a: :::;
0 (f) 0
<(
~m II
C)
U.J
) '
..
\
o\I '
'
I
SELECTED CASE STUDIES 273
Static water levels are typically about 11 ft below land surface, which
is at a mean sea level elevation of about 44 ft. During recharge, interfer-
ence between wells is such that wellhead pressures can reach about 28 psi
while during recovery, drawdowns of up to about 20m (65ft) can occur
in individual wells.
Following expansion of the ASR system from one to six wells, a system
performance test was performed, recharging all six wells at once and
monitoring differences in recharge rates and volumes and recovered water
quality from each well. Extensive data were collected, some of which is
presented in Table 9.2. It is pertinent to note in this table the relative water
quality at the end of recovery between Well R-1, which had been in ASR
operation for four years, and the remaining wells that were being placed into
service. The improvement in quality with successive cycles is apparent.
274 GROUNDWATER RECHARGE AND WELLS
Since ASR operations at all six wells began in 1991, data have been
gathered to monitor system performance. Figure 9.4 shows monthly and
cumulative storage volume for all wells for the period September 1992, to
August 1993, and variations in recharge and recovery water quality, as
indicated by chloride concentrations. Performance is satisfactory except
for Well R-2, which consistently shows higher turbidity compared to the
other wells. This continues a pattern of abnormal response in Well T-2 that
began during well construction and testing. Such differences in well
performance in karst limestone aquifers are not unusual. Plugging has not
been observed during ASR operations to date.
The target storage volume for each ASR well is 100 MG (378 megaliters)
per MG/day recovery capacity, thereby providing about 100 days of
recovery. Full recovery efficiency has been demonstrated in this brackish
aquifer. Considering that the seasonal variability in supply from the wellfield
is small, the only real variability is in demand. Such a target storage
volume should enable the city to meet seasonal demand variations up to a
peak demand of 212 megaliters/day (56 MG/day).
Further increases in demand will be met from a new surface water
source, Taylor Creek reservoir, which is located about 3 miles from the
water treatment plant. The supply of water from this source is highly
variable, reflecting regulatory and environmental restrictions upon
streamflow diversions. In many years, periods of no diversion will extend
for about 1 month, while a period of up to about 18 months is possible. As
a result, the ASR target storage volume associated with this source will be
substantially greater than 378 megaliters (100 MG) in order to meet
seasonal and long-term variability in both supply and demand.
SELECTED CASE STUDIES 275
109
100
M A
88
100
* Florida Keys Aqueduct Authority, 1100 Kennedy Drive, Key West, Florida 33041
276 GROUNDWATER RECHARGE AND WELLS
Subsequently, it was determined that the well was not fully developed.
Additional development was conducted, improving well specific capacity
to about 2.7 G/min/ft at typical pumping rates of 235 to 350 G/min. No
subsequent hydraulic testing has been performed to reevaluate hydraulic
characteristics or their response to several cycles of ASR operation. No
signs of well plugging have been observed.
The ASR well and two observation wells are constructed as shown in
Figure 3.1. 400 mm (16 inch) Schedule 80 PVC casing was cemented to
a depth of 118m (387 ft). A 250 mm (10 inch), 0.025 slot stainless steel
316, wire-wrapped screen was then installed to a depth of 130m (427ft),
plus a 1.5 m (5 ft) sump. The screen was extended with a blank section
Collier County I
---~'--·- - - - : - : : '
,_....,
Scale in Miles
;
0 10 20
General
Locati~
Gulf of Mexico
Atlantic Ocean
inside the casing to a depth of 110 m (362 ft) and the annular space was
filled with 6/20 gravel using a l-inch gravel tube.
During well construction, continuous wireline cores were collected
between 107 and 137m (350 and 450ft). The storage zone is composed
of clean, coarse quartz sands with a porosity of 26 to 35% with minor
amounts of carbonate, calcite, and dolomite and traces of smectite, illite/
mica, and kaolinite. These cores were analyzed to determine their physical,
geochemical, and other properties to provide a basis for ASR zone
selection and screen design. From this information and associated geo-
physical logs, observation well OW -1 was equipped with three 0.5-inch
sampling tubes, isolated with packers into production intervals at the top
(387 to 405 ft), middle (405 to 418 ft), and bottom (418 to 428 ft) of the
storage zone. Sample ports are set at depths of 122, 125, and 130m (400,
410, and 426 ft).
The storage zone contains seawater. An observation well core descrip-
tion of the storage interval and adjacent portions of the overlying and
underlying confining layers is shown in Table 9.3. Figure 9.6 shows the
water quality response at observation well OW -1 during Cycle 2. It is of
some interest that the freshest water occurs at the bottom sampling interval
during the recharge, storage, and most of the recovery period. During
Cycle 2, 37 megaliters (9.7 MG) were injected over a 42-day period. The
water was stored for 34 days and then 13 megaliters (3.5 MG) were
recovered. Recovery efficiency was about 23% during this cycle before
water quality exceeded potable standards. Evidently the vertical hydraulic
conductivity within the storage zone is sufficiently low that density strati-
fication occurs quite slowly. In subsequent cycles, recovery efficiency
steadily improved as the storage zone was developed, reaching about 72%
during the test program.
Static water level in the ASR well typically ranges from 0.9 to 1.2 m (3
to 4ft) below the measuring point, which is close to land surface elevation
of about 1.5 m (5 ft). During recharge and storage periods, freshwater
displaces seawater in the storage zone, increasing static head above land
surface. Daily tidal variation is about 0.1 m (0.4 ft).
A series of eleven ASR cycles has been conducted to determine the
relationships between recharge and recovery rates, volumes, storage peri-
ods, and recovery efficiencies. Table 9.4 shows water quality of the
recharge water and the native groundwater at this site. The storage zone
TDS is 37,200 mg/L. Simulation modeling was conducted prior to the test
program to provide a basis for design of the test cycles.
From the results of the cycles, it is apparent that treated drinking water
can be stored in a thin, confined, low permeability aquifer containing
TABLE 9.3 CORE DESCRIPTION: MARATHON, FLORIDA
CR6 378.6-383.6 5.0 5.4 108% Light olive Sandstone vf grained in 5ft in 11/ 2 min.
calcareous clayey matrix,
well consolidated little-
same pelecypod shell
(10y5/4) subrounded-Moldic porosity
throughout sample. Trace
medium coarse silica.
CR7 383.6-388.6 5.0 3.9 77% Light olive As above; slightly more 1 ft ss; 11/ 2 soft,
(10y 5/4) phosphorite grains in 1/2 ss, 2 soft.
and yellowish lime-mud matrix. Middle 11/ 2 soft
Gray (5y 7/2) Coarse-grained silica has some hard lenses
angular-rounded. Well less than .05 ft
rounded calcareous sand thick. Sand sample
grains (fine-grained). from core-catcher
obtained 5 ft in.
11/ 2 min.
CR8 400-405.5 5.5 2.4 44% Light olive Sand, vf-med, poorly 5 1/ 2 ft in 21/ 2
(10y 5/4) sorted silica slightly min. 1-ft sample to
effervesent in HCL Mineralogy Inc. for
angular-subangular. analysis.
Little vf. black
phosphorite in matrix.
Little white (N9)
limestone fragments in
upper 0.6 ft of core.
Some lime-mud matrix.
CR9 406-410.5 4.5 3.2 71% As above As above; limestone frags No core return from
occur throughout core. 405.5-406, 4.5 ft
Iron staining but may be in 3 1/ 2 min.
from core barrel.
CR10 411-413.5 2.5 2.4 96% As above As above - with gravel 2.5 ft in 5 mins.
(<.4mm) Subangular to 2-ft sample to David
subrounded. Pyne.
CR11 413.5-416.5 3.0 3.3 110% As above As above More coarse with depth
3.0 ft in 5 min 40 sec.
Mud viscosity = 45 sec.
Mud weight= 8 3/4 lb/gal.
1 ft sample to Mineralogy
Inc.
CR12 416.5-417.4 0.9 0.9 100% As above As above Lost circulation. Retrieved
core barrel which
has this sample in it.
CR13 417.4-420.4 3.0 2.0 67% As above As above. Appears finer Sample 420-420.4.
grained. Less phos-
phorite, calcareous matrix,
mottled streaks.
CR14 422-427 2.0 1.8 90% As above As above. Less calcareous Mud viscosity = 60 sec.
and pale material subrounded-rounded. Mud weight= 83/ 4 .
olive (1 Oy 6/2) Predominantly translucent Casing set but it just
silica grain. Poorly was set down into sand.
sorted. Core obtained most
likely from 424'-427'.
5 ft in 4 min.
CR15 427-428.5 1.0 0.0 0% As above As above Switch to conventional
shoe for next run
because we're
in hard material
at 428.5 (may be clay
lens).
CR16 428.5-433.5 5.0 1.8 36% As above As above Core indicates a cleaner
sand below the olive
sand we've encountered
thus far. Bottom of
core is very hard
sandy clay.
CR17 433.5-438.5 5.0 4.7 94% Light olive As above; thin (<2") Predominantly sand with
(10y 5/4) clay lenses @434' thin clay lenses.
olive and gray and 436'-438'. Clay
(5y 3/2) is highly plastic (CH).
280 GROUNDWATER RECHARGE AND WELLS
25,000
I -----Top
Tes1Cycle2 Middle
20.000 ~ - - - - - - Bottom
I I
::::;
-~
~ ~
Nota: Observation well located
126ft. from ASR well. ASR storage
interval is 387~427 ft.
0; 15,000
g '
\~ I I
I
I
Recharg Storage Recovery
Ql
:g
~ 10,000
\'
-~ T 1 I
u "7
~\ .I
5.000
\
\
~
--, ~~
--::::..:- 1-, v
0
-- -~
0 10 20 30 40 50 60 70 80 90 100
Time (days)
Figure 9.6 Observation well chloride concentration vs. time, Marathon, Florida.
TABLE 9.4
RECHARGE AND NATIVE WATER QUALITY: MARATHON, FLORIDA
pH 10.3 7.60
Total alkalinity 23.1 120
Conductivity (!lmhos/cm) 397 49,000
Carbonate hardness 110 1,390
Non-carbonate hardness 95.0 6,480
Turbidity (NTU) <0.2 <0.2
Total dissolved solids 212 37,200
Total suspended solids <1.0 4.2
Calcium 33.8 398
Magnesium 3.75 1,250
Sodium 20 11,000
Potassium 11.4 385
Silica 4.7 9.43
Aluminum <0.5 <0.5
Iron 0.05 <1.0
Chloride 41.8 20,800
Fluoride 0.8 0.84
Sulfate 91 '1 2,910
Nitrate and nitrite <0.02 <0.02
Carbonate 16.8 0
Bicarbonate 23.1 146
Note: Recharged water sampled April 3, 1990. Native water sampled May 4,
1990.
SELECTED CASE STUDIES 281
* Town of Palm Bay, 1101 Troutman Boulevard, NE, Palm Bay, Florida 32905
SELECTED CASE STUDIES 283
The static water level during the test program was a few feet below land
surface, which is about 15 ft above sea level. The storage zone is a
production interval at the top of the Ocala formation, confined above by
clays of the Hawthorn formation and below by low permeability lime-
stones. Construction details of the ASR well, a production well in the
overlying secondary artesian aquifer, and ASR observation wells are
shown in Figure 9.7. The ASR well is constructed of 300 mm (12-inch)
PVC casing to 91 m (298ft), and is open hole to 113m (370ft). The two
production zone observation wells are located at distances of 15 and 30 m
(50 and 100 ft) from the ASR well while the deep monitor well in an
underlying production zone is located 12m (39ft) from the ASR well. The
well is equipped with a vertical turbine pump and a chlorinator, and
recharge occurs down the annulus.
Storage zone native water quality constituents of interest included TDS
= 1320 mg/L, chloride = 588 mg!L, and hydrogen sulfide = 3.4 mg/L.
Recharge water chloride concentration averaged about 180 mg/L. Since
the drinking water standard for chloride is 250 mg/L and the ASR well is
located within the distribution system, the opportunity for blending the
recovered water with treated water from the wellfield is limited.
Recharge occurred at variable rates between 1.1 and 3.8 megaliters/day
(200 and 700 G/min), depending upon flow available from the water
treatment plant. Recovery also occurred at variable rates, ranging between
1.4 and 4.1 megaliters/day (260 and 750 G/min), depending upon system
operational requirements. Figure 9.8 shows the improvement in chloride
concentrations occurring during the first three ASR test cycles, which
included small storage volumes of about 13 megaliters (3.5 MG). The first
cycle recovered 170% of the injected volume, in order to trace the initial
mixing characteristics for this well. The second and third cycles recovered
water until chloride concentrations reached about 275 mg/L, leaving a
small buffer zone underground to enhance recovery efficiency in subse-
quent cycles.
Based upon the successful results of these cycles, two larger cycles were
then conducted. In cycle 4, 159 megaliters (42 MG) were injected; how-
ever, only 61 megaliters (16 MG) were recovered due to low demands in
the water system at this time. The balance was left in the aquifer to enhance
recovery efficiency in subsequent cycles. Chloride concentration at the
end of recovery was 196 mg!L. In cycle 5, 227 megaliters (60 MG) were
injected and 163 megaliters (43 MG) were recovered before chloride
concentration reached about 250 mg/L. At 100% recovery, chloride con-
centration reached 310 mg/L, compared to 588 mg!L for the native water
in the aquifer.
N
co
.j:>.
PZ-1 and PZ-2 R-1 DMW-1
~=~~i~~
Q i
D(r,'h 1 F Ia;;:& if 1 I PI <011 1-' • ..._, '""',. ""'' ,.. , ""'"' ••••J
100
200
300
G>
JJ
0
c
z
400 Cl
~
m
JJ
500 JJ
m
0
I
:l>
JJ
G>
600 m
>
z
Cl
::§:
m
r
r
Figure 9.7 Construction details of the ASR facility, Port Malabar, Florida. (j)
SELECTED CASE STUDIES 285
&n.-------------------------------------------,
Backgrouf19 Chloride 588 mg/L
&n ~~~e~n~d--------~----~~~~--~
- - Cycle No. 1. 3.47 MG Injected, 5.89 MG Recovered
- - - • Cycle No.2, 3.37 MG Injected. 2.36 MG Recovered
- · - Cycle No.3, 3.48 MG Injected. 3.00 MG Recovered
400
10 20 30 40 50 60 70 80 90 100
Percent Recovery
water system was reduced by about 30% due to loss of a major industrial
water user that changed its water source to onsite wells. With the reduction
in system demand, no recharge or recovery occurred for about three years
other than the trickle flow. Rapid growth in water demand then com-
menced, precipitating the need for recovery of the stored water. However,
recovery efficiency was less than expected. Analysis of available data
suggests that the redistribution of pumping in the area probably pulled
some of the stored water away from the ASR well towards the industrial
supply wells, which are located about 1859 m (6100 ft) from the ASR site.
Neglecting dispersion, the theoretical radius of the stored water bubble
around the ASR well is about 488 m (1600 ft). With seasonal operation as
originally planned, it is unlikely that any reduction in recovery efficiency
would have been noted. However, long term storage in this zone for
periods of several years may be at the expense of some loss in recovery
efficiency. An alternative would be to use a deeper producing interval at
a depth of 141 to 163 m (464 to 534ft) for ASR storage at such time as
the system may be expanded.
* City of Boynton Beach, 124 SE 15th Avenue, Boynton Beach, Florida 33425
SELECTED CASE STUDIES 287
During the first three ASR test cycles, recharge and recovery specific
capacities were similar, ranging from 13 to 19 G/min/ft.
Static water level prior to the beginning of ASR cycles was about 14m
(45ft) above the measuring point elevation, which was at 6 m (19ft) above
mean sea level. As the storage zone around the well was alternately
displaced with freshwater and brackish water during ASR cycle testing,
the density of the water column in the well varied. Consequently, the static
water level rose in the well during recharge and declined during recovery,
to a greater extent than would have occurred without density changes.
During recharge, the injection pressure reaches about 28 m (93 ft) above
mean sea level while during recovery, water levels reach about 8 m (27ft)
below mean sea level.
The storage zone is characterized as yellowish and pinkish grey, fossil-
iferous biomicritic limestone of Oligocene age, overlain by calcareous
clay, fossiliferous limestone, chert, and shell. The overlying sediments and
the underlying limestone are not productive.
The ASR well is designed with 400 mm (16-inch) diameter steel casing
to a depth of 245 m (804 ft), below which the well is open hole to a depth
of 277 m (909 ft). Originally it was drilled to a depth of 1260 ft to
determine the location and quality of water from potential deeper storage
intervals. Once the ASR interval was selected, the hole was plugged back
to its final depth.
Large quantities of fine sand were pumped from the well during an
extended period of airlift development; however, concentrations showed a
reducing trend. Sand production ceased when the well was subsequently
equipped with a vertical turbine pump.
Recharge occurs down the annulus at flow rates of up to about 5.4
megaliters/day (1 000 G/min). Recovery occurs at about 3.2 to 5.4 megaliters/
day (600 to 1000 G/min). Table 9.5 shows typical recharge and recovery
water quality. The storage zone is a semi-confined limestone artesian
aquifer with a native water TDS concentration of 3910 mg/L.
Cycle volumes have not exceeded 227 megaliters (60 MG). Backflushing
occurring at the beginning of recovery has been sufficient to maintain
recharge specific capacity at original values. During the first seven cycles,
recovery efficiencies at this volume have reached 80% and are showing
steady improvement. Table 9.6 shows the volumes, flow rates, recovery
efficiency and specific capacity results for the first three cycles. Despite
considerable effort, the reason for the unusual pattern of variation in
TABLE 9.5 RECHARGE AND RECOVERY WATER QUALITY: BOYNTON BEACH, FLORIDA
Chlorine Residual
Hardness Organ.
Alkalinity Turbidity CL Free Total NH3 Sulfate Iron Sodium TSS TDS DO THM
pH "T" "Total'' "CA'' (NTU) (mg/L) (mg/L) (mg/L) (mg/L) Color (mg/L) (mg/L) (mg/L) (mg/L) (mg/L) (mg/L) (!!gil)
Background 7.60 130.00 930.00 94.50 0.00 1920 NA NA 0.73 0.00 436.00 0.02 3910.00 NA
Cycle 1
Recharge
Minimum 8.80 29.00 80.00 73.00 0.03 49 2.50 3.00 0.00 1.00 30.50 0.00 17.00 1.00 156.00 9.50 152.00
Maximum 9.60 47.00 122.00 100.00 0.10 150 4.00 4.20 0.00 1.00 30.50 0.00 17.00 23.00 178.00 10.70 170.40
Average 9.06 35.64 92.07 85.14 0.08 58 3.25 3.60 0.00 1.00 30.50 0.00 17.00 12.00 167.00 10.10 160.37
----
Cycle 1
Recovery
Minimum 8.10 33.00 102.00 93.00 0.07 57 0.00 0.00 0.00 2.00 42.90 0.05 42.00 0.00 288.00 2.00 113.00
Maximum 8.90 80.00 462.00 369.00 0.19 756 3.50 0.00 0.20 5.00 206.00 0.45 382.00 23.00 1422.00 6.20 202.00
Average 8.39 53.63 237.00 198.63 0.12 168 1.75 0.00 0.06 3.60 103.60 0.19 212.00 11.50 855.00 3.53 162.33
- --
Cycle 2
Recharge
Minimum 8.50 28.00 79.00 72.00 0.03 44 1.50 4.10 0.10 2.00 22.60 0.00 16.50 0.00 190.00 4.90 0.00
Maximum 9.50 77.00 132.00 130.00 0.70 72 5.00 5.90 0.35 11.00 28.70 0.00 21.70 5.00 329.00 5.50 0.00
Average 9.08 41.17 93.86 87.50 0.11 51 3.50 4.83 0.14 7.29 25.77 0.00 19.58 1.00 229.86 5.20 0.00
----
Cycle 2
Recovery
Minimum 8.00 39.00 86.00 80.00 0.00 50 0.00 0.00 0.52 3.00 40.00 0.05 10.00 0.00 349.00 1.10 69.60
Maximum 9.20 49.00 246.00 207.00 6.64 424 0.00 0.20 1.60 4.00 58.60 0.06 118.50 6.00 598.00 7.50 72.10
Average 8.62 44.50 161.63 141.75 0.56 255 0.00 0.03 1.34 3.67 47.53 0.06 56.50 3.00 450.67 4.30 70.85
----
Cycle 3
Recharge
Minimum 8.50 32.00 74.00 68.00 0.04 40 0.00 3.70 0.10 6.00 22.15 0.00 20.34 0.00 161.00 5.90 0.00
Maximum 9.40 77.00 112.00 104.00 1.65 52 0.00 5.10 0.20 10.00 27.55 0.00 22.00 0.00 263.00 5.90 0.00
Average 9.11 45.81 91.10 85.21 0.14 46 0.00 4.85 0.13 8.00 24.71 0.00 20.98 0.00 205.17 5.90 0.00
----
Cycle 3
Recovery
Minimum 8.30 46.00 115.00 103.00 0.11 50 0.00 0.00 0.22 3.00 24.77 0.00 21.00 0.00 233.00 0.00 87.00
Maximum 8.90 66.00 228.00 199.00 1.42 300 0.00 0.00 1.02 6.00 38.60 0.00 96.40 4.00 667.00 0.00 87.00
Average 8.43 53.17 162.00 141.43 0.21 162 0.00 0.00 0.56 4.80 30.69 0.00 54.33 2.00 440.00 0.00 87.00
Note: 1. 0.00 means the parameter was EITHER not analyzed OR not detected above method detection limits. 2. NA: not applicable. 3. Temperature and conductivity measured but not shown in this table. Cycle 1 recharge
total organic carbon averaged 15.3 mg/L against a background of 1. 7 mg/l. Temperature and conductivity measured but not shown in this table. Cycle 1 Recharge total carbon averaged 15.3 mg/L againsl a background
of 1.70mg/l.
SELECTED CASE STUDIES 289
TABLE 9.6
ASR CYCLE TESTING SUMMARY: BOYNTON BEACH, FLORIDA
Recharge/
Recharge Recovery Specific
Total Recovery Rate at End Capacity Recovery
Duration Volume of Cycle At End Efficiency
Cycle (days) (G) (G/min) (G/min/ft) (%)
Cycle 1
Recharge 14 12,525,510 720 19
Recovery 8 9,578,934 934 13 55
Cycle 2
Recharge 44 57,323,180 940 15
Recovery 31 26,099,896 968(a) 15 42
Cycle 3
Recharge 43 58,342,670 930 17
Recovery 25 32,240,565 1,032 17 50+(b)
(a) Recovery during Cycle 2 was performed over a range from 330 to 1,000 G/min
to estimate water quality sensitivity to recovery rate.
(b) Recovery terminated at chloride concentration of 270 mg/L compared to 350
mg/l on previous cycles.
recovery efficiency during the first two cycles (55 and 42%, respectively)
was not determined. Recovery efficiency is expected to continue increas-
ing in subsequent cycles. The target water quality during recovery was a
TDS of 350 mg/L.
The objective of the ASR system at this site is to augment seasonal
water supplies. The city currently has sufficient treatment capacity; how-
ever, raw water supplies are limiting, reflecting declining capacity of
existing wells and difficulties in locating suitable new well sites in this
urban area. Recovery of stored water from the ASR well during peak
demand months provides a significant benefit to the city. Consideration is
being given to future construction of a second ASR well in the distribution
system, to augment peak supplies from the wellfield and also to resolve
distribution system hydraulic constraints.
* South Florida Water Management District, PO Box 24680, West Palm Beach, Florida 33416
290 GROUNDWATER RECHARGE AND WELLS
was from Taylor Creek-Nubbin Slough, which drains a dairy farming area.
Efforts were therefore initiated by the South Florida Water Management
District to improve the condition and operation of Lake Okeechobee.
Among these efforts, the District investigated ASR as a method for storing
seasonal peak flows from Taylor Creek-Nubbin Slough in a brackish
aquifer, recovering the water to meet agricultural irrigation needs and
keeping it out of the lake.
The ASR test well and a dual-zone observation well are adjacent to the
L-63N Canal near the town of Okeechobee. Canal water quality during the
testing period varied from TDS values of 268 mg/L to a high of 996 mg/
L, depending upon local rainfall. Water is obtained from the canal and is
pumped into a shallow basin for preliminary settling and to provide some
chlorine detention time, if chlorination is needed. It is then repumped
down the ASR well at a design rate of 19 megaliters/day (5 MG/day). The
shallow basin can be bypassed, if desired. Well hydraulics are such that
recharge flows at double this rate could easily be achieved with the
appropriate pumping equipment. Recharge occurs in the upper Floridan
Aquifer System at depths of 366 to 518 m (1200 to 1700 ft). The water is
recovered by gravity flow at initial rates of about 17 megaliters/day (4.5
MG/day), declining to about 13 megaliters/day (3.5 MG/day) as the den-
sity of the water column in the well increases and as recovery hydraulic
effects approach equilibrium. ASR well testing began in May 1989 and
continued intermittently until September 1991.
An ASR well pumping test indicated the following characteristics for
the Upper Floridan Aquifer:
ASRWeU
Depth
(fl)
0
Undifferentiated deposits
of sands, clay, and shell Surficial 100
Aau~er
·t.-:::.P.: 200
Sandy phosphatic clay
with interbedded shell Confining
Units and ·:.·:::.1!:
Minor 300
Water .y.-::::.
Yield
Zonas 400
700
~r~~~~i':d '~n';_l~~~J~~~s
limestone; abundant 800
Lepido;::,yclina present
900
Pale orange, poorly
consolidated, Artesian
fossiliferous limestone Aquifers 100
Abundant
Lepidocyclina and 1100
Dictyoconus present
1200
1300
1400
Dark yellowish brown,
hard, porous, fine 1500
crystalline dolomite
1600
1700
T.D. 1700'
1800
Le and
!·.·.·.·.·J Sand E!2i3 Fossiliferous Limestone
~ Sand & Shall Fragments ~ Dolomite Limestone
~Clay P Phosphorite
~ Calcareous Clays
The ASR well is constructed with 600 mm (24-inch) black steel casing
to a depth of 386m (1268 ft), below which the well is open hole to 521
m (1710 ft). Recharge occurs down the well casing, which has no pump,
injection tubing, or other internal piping. Reflecting the high transmissiv-
ity of the storage zone, no well plugging has been experienced during
testing to date.
292 GROUNDWATER RECHARGE AND WELLS
During initial ASR cycle testing, small volumes were stored for short
periods to ascertain the effect of the brackish water storage zone on the
reduction of coliform concentrations in the recharge water. In general, it
was shown that coliform concentrations reduced to acceptable levels
within 1 to 3 days of storage in the brackish aquifer. Chlorination of
recharge flows was not found to be necessary to protect water quality,
public health or welfare, and was also not necessary to prevent well
plugging due to bacterial activity, reflecting the high transmissivity of
this well.
The ASR well was designed to achieve maximum recharge rates at the
expense of recovery efficiency. The combination of substantial storage
zone thickness, high transmissivity and poor native water quality is a
difficult combination if the objective is to achieve high recovery efficiency
in an ASR well. Test cycles completed to date have shown recovery
efficiencies increasing to 35%, associated with steadily increasing storage
volumes of up to 1.9 Mm 3 (500 MG). Under normal seasonal operation,
larger storage volumes would be expected. Improvement in quality with
successive small volume ASR test cycles at this site suggests that larger
volumes would probably achieve recovery efficiencies in the range of 40
to 60%, or possibly higher. Plugging back the well to eliminate productive,
very brackish intervals at the bottom of the well would improve recovery
efficiency but would probably reduce potential recharge and recovery
rates.
While this recovery efficiency may appear to be quite low, further
analysis has shown that any recovery efficiency greater than about 40% is
a net gain to the regional water management system. Evapotranspiration
and seepage losses for water stored in Lake Okeechobee and the adjacent
Water Conservation Areas, or conveyed from the lake to urban areas
through the canal system, range from about 60% near the lake, to at least
90% during conveyance to more distant parts of the region. Furthermore,
water stored but not recovered may serve as a source of recharge to the
brackish upper Floridan Aquifer System, which is being developed as a
regional water source for brackish water to supply reverse osmosis water
treatment plants along the coast. Over a period of several decades it is
expected that an expanded ASR system recharging canal water, either
directly or from adjacent shallow wells, would recharge the Upper Floridan
Aquifer and also tend to regionally reduce TDS concentrations.
The original objective of the ASR facilities at this site was to divert as
much phosphorus-laden water as possible out of Lake Okeechobee. A
secondary objective was to recover water for agricultural irrigation pur-
poses during the dry season. Economic analyses conducted during 1989
SELECTED CASE STUDIES 293
Between 1988 and 1990, the City of Chesapeake, VA, completed con-
struction and initial testing of an ASR well, two observation wells, a
pipeline, an access road, wellhead facilities, a disinfection system, and a
wellhouse at the city's Western Branch wellfield in the northwest comer
of the city's water service area. Recharge water is obtained from the
Northwest River water treatment plant which has a capacity of38 megaliters/
day (10 MG/day). Due to movement of saltwater up the estuary, the
Northwest River is not available as a reliable water source during dry
summer months. At such times, water supply has historically been a
challenge for the Chesapeake area, requiring greater reliance upon whole-
sale water purchases from surrounding utility systems. Seasonal storage of
available water supplies during wet months was seen as a potentially
viable and cost-effective water supply alternative.
Facilities include an ASR well equipped with a 19 megaliters/day (5
MG/day) vertical turbine pump. The casing is 600 mm (24 inch) in
diameter and constructed of epoxy-coated steel to a depth of 119 m (390
ft). The storage zone includes the Upper and Middle Potomac aquifers of
Cretaceous Age. These are clayey sand aquifers in the Atlantic Coastal
Plain. The upper screen section is from 171 to 189m (560 to 620ft), while
the lower section is from 203 to 221m (665 to 725ft). The total well depth
* Department of Public Utilities, City of Chesapeake, PO Box 15225, Chesapeake, Virginia 23320
294 GROUNDWATER RECHARGE AND WELLS
100
200
a; 300
g
ID Formation K Packer
0 380' to 390'
E
':;
<n
"0 24' Casing
c 390' to45Q'
::> 400
0
0
3
0
a;
C)
<:
o_
(])
0 500
800
is 227 m (745 ft). Figure 9.10 shows the well construction. Operation
began in 1990.
Aquifer hydraulic characteristics were determined from pumping tests,
with the following representative results:
Leakance negligible
Specific capacity 31 G/min/ft at 1250 G/min (recharge, Cycle 1)
36 G/min/ft at 2200 G/min (recovery, Cycle 1)
Groundwater
Parameters 570-585 ft 675-690 ft Recharge Water
300,-----------------------------------------------~
50 vA·'"·"-"'·"-"·"-"·"·'"·'"·"-"-"-"'-""-"'-""-"-"-""--:::.:.-..~.~-~---------------------
I I I I I I I I I
Q~0--~10~--~~----~W----4~0----5~0----6L0----7L0--~8~0--~9-0--~100
water subsequent to the test program was considerably lower than values
recorded during the test program. The lower pH mobilized manganese
present in the aquifer, causing discoloration of recovered water. A pH
adjustment system was installed to ensure that recharge water pH remains
at least above 8.0. Some of the stored water was pumped out to remove
more acidic waters with associated concentrations of manganese. A trailer-
mounted manganese removal treatment system was then tested to confirm
that any ASR recovered water containing manganese could be treated prior
to pumping into the distribution system. The city has continued to recharge
the ASR well with water at acceptable pH. Recovered water quality has
shown steady improvement with successive cycles, suggesting that pH
adjustment is probably sufficient for controlling manganese concentra-
tions once the residual acidity in the aquifer from earlier low pH operations
has been overcome.
This experience suggests that pH control is particularly important in
clayey sand aquifers, even when core analyses and geochemical analyses
suggest that no problems are expected. The difference is due to changing
pH values, outside the range of the original analyses. The ten core intervals
selected for detailed analysis apparently did not include significant con-
centrations of manganese-bearing minerals that were present in the aqui-
298 GROUNDWATER RECHARGE AND WELLS
* New Jersey American Water Company Eastern Division, 100 James Street, Lakewood, New Jersey
08701
SELECTED CASE STUDIES 299
Depth
Below
Grade Legend
568~;:~
~
Famolbn § CLAY
Wooa:.sy
c~
~
Bed
~ 580 E1 SILT
Famolbn
0 fine to medium SAND
.
lJppS<
Pi1M
AQJ;I<M 590 ~ coarse SAND
~ LIGNITE
~'EPIDOTE
§] CHLORITE
~ SULFUR
II1l BURROWS
[;3 PYRITE
~ Rorltm
-
Bed Fotmatlol1
30-
Cement
470- grout from
592 teet to
grade
570-
Marie #3 grade
650- sand filter pack
670-
TABLE 9.8
RECHARGE WATER QUALITY: SWIMMING RIVER, NEW JERSEY
NJDEP Drinking
November 1988 January 1990 Water Standards
Constituent (mg/l) (mg/l) (mg/L)
pH 8.2 8.7 6.5-8.5
Total dissolved solids 212.0 500
Hardness 68.0
Alkalinity 38.0
Bicarbonate 30.0
Calcium 21.21 41.0
Chloride 37.43 3.85 250
Fluoride 1.0 2.0
Nitrite <0.01
Magnesium 3.67 5.00
Nitrate as N <1.0 2.27 10
Sodium 27.65 25.60
Sulfate 37.0 38.40 250
Pesticides <0.0001
Aluminum 0.059 1.00
Antimony <1
Arsenic <0.005 0.05
Barium <0.05 1.0
Beryllium <0.1
Boron <0.1
Cadmium <0.0002 0.010
Chromium <0.001 0.05
Cobalt <0.1
Copper <0.02 1.0
Iron <0.05 0.12 0.3
Lead <0.005 0.05
Manganese <0.02 <0.01 0.05
Mercury <0.005 0.002
Molybdenum <0.5
Nickel <0.05
Potassium 3.28 3.50
Selenium <0.003 0.01
Silver <0.0002 0.05
Strontium <0.2
Thallium <0.5
Vanadium <5
Zinc <0.055 5.0
Total THMs 0.041 0.095 0.10
Silica 5.96
Phosphate 0.06
NJDEP Drinking
June 1989 January 1990 Water Standards
Constituent (mg/L) (mg/l) (mg/L)
was no longer present around the well, the pyrite in the aquifer had
probably been "polished" by the acidization pretreatment process and was
therefore highly reactive to oxygen-bearing water, forming a ferric hy-
droxide precipitate. The iron content of the recovered water reduced with
successive cycles, reflecting a probable buildup of oxidation products on
the pyrite grains, thereby reducing pyrite reactivity. However, the iron
concentrations in the recovered water were still too high. As a result, pH
304 GROUNDWATER RECHARGE AND WELLS
* Wildwood Water Utility, City of Wildwood, Pine and New Jersey Avenue, Wildwood, New Jersey
08260
SELECTED CASE STUDIES 305
(80 MG) is recovered from each well, while the remainder is left under-
ground to help control saltwater migration into the aquifer.
Recharge rates in individual wells typically range from 2.2 megaliters/
day (400 G/min) at the beginning of the recharge season to about 1.1
megaliters/day (200 G/min) at the end, while recovery rates are about 3.5
to 5.4 megaliters/day (650 to 1000 G/min). Summer water demands aver-
age about 38 megaliters/day (10 MG/day) while peaks reach 49 megaliters/
day (13 MG/day). About 13 megaliters/day (3.5 MG/day) of the peak
summer demands are met from the ASR wells. Winter demands average
about 9 megaliters/day (2.5 MG/day). The ASR wells provide system
reliability in case the pipeline from the interior wellfield is broken acciden-
tally. The wells also provide distribution system storage, which has been
partially substituted for elevated storage in the coastal service area.
All four ASR wells utilize the Cohansey aquifer. Static water levels in
this aquifer are about 4 m (12ft) below land surface, while well depths are
about 76 m (250 ft).
Aquifer hydraulics are not well known. Transmissivity of one well in
the Cohansey aquifer has been estimated at about 1,080 m2/day (87,000 G/
day/ft). This well has a total depth of 100m (328ft) and has a 300 mm (12-
inch) screen (0.045 slot) from 65 m (212 ft) to the bottom. During a
pumping test at a rate of 5.6 megaliters/day (1034 G/min) the specific
capacity was 24 G/min/ft. Recovery specific capacities typically range
from 9 to 13 G/min/ft, associated with drawdowns of about 30m (100ft).
The wells are backflushed daily for about 10 min to remove iron floc
formed from reaction of oxygen and chlorine in the water with ferrous iron
present in the aquifer, and also to remove rust from the distribution system,
which enters the wells during recharge. The wells are fully redeveloped
about every 5 years and are chemically treated with surging and acid
treatment about twice per year to maintain the screen capacity.
The ASR system has operated successfully for 26 years and two more
ASR wells are planned.
* Upper Guadalupe River Authority, 215 W. Water Street, Kerrville, Texas 78029
306 GROUNDWATER RECHARGE AND WELLS
ing the summer at typically 1.53 times average demand. The Authority
provides water at rates up to the plant capacity, and the city meets higher
summer peak demands from its wells.
Faced with the need to construct a new offstream reservoir at great
expense to meet projected increases in water demand while also meeting
downstream minimum flow requirements, the Authority conducted an
investigation of ASR feasibility. This investigation was completed suc-
cessfully in three phases between 1988 and 1992, including a preliminary
feasibility investigation; construction and testing of an ASR zone observa-
tion well; and construction and testing of an ASR well, wellhead facilities,
and two additional observation wells in overlying production zones. The
associated facilities are operational and were fully permitted during 1993.
The ASR storage zone is the Hosston-Sligo formation in the Lower
Trinity aquifer. The storage zone is a confined sandstone and conglomer-
ate artesian aquifer. Hydraulic testing at the ASR well indicated the
following aquifer characteristics:
During the test period, static water elevation was about 460 m (1510 ft)
above sea level, or about 39 m (128 ft) below land surface elevation of~
about 499 m (1638 ft). Local gradient of the potentiometric surface was
about 0.0011. Table 9.10 shows the general lithology of the storage zone
and overlying formations.
Average Thickness
in Kerrville Area
Formation Name (ft) Primary Materials
would continue to supply water at an acceptable rate for the 4-year dura-
tion of the drought. This drought provides an important reference point for
water supply regulation in Texas. The storage volume required for Kerrville
is about 4.3 Mm3 (3500 acre ft) to meet projected demands in the year
2040.
With the success of the ASR program, the city was able to save the
approximately $30 million expected cost of an offstream reservoir. In-
stead, ASR wells can be constructed in low-cost increments to meet
increasing demands, at a total expected cost of under $3 million. To meet
projected needs up to the year 2015, an investment of about $1 million is
anticipated.
This project won an Honors A ward in the American Consulting
Engineers Council National Engineering Excellence Competition for
SELECTED CASE STUDIES 309
1993. Since then, the City of Kerrville has converted an existing water
supply well, R-5, to ASR operation, supplementing the yield of Well
R-1.
This site has probably the lowest transmissivity of the operational ASR
sites in the U.S. as of 1993.
The depth to static water level during the test program was about 250
to 268 m (820 to 880 ft). Less than 1% of the mean annual precipitation
of 279 to 457 mm (11 to 18 inches) reaches the Arapahoe formation as
* Centennial Water and Sanitation District, 62 West Plaza Drive, Highlands Ranch, Colorado 80126
310 GROUNDWATER RECHARGE AND WELLS
660
680
700
720
., 740
760
~ 780
.! 800
~
,!,l 820
:E 840
c.
GO
a 860
880
900
920 Depth of Uppermost Water-Bearing Zone
940
960
980
1000
1985 1986 1987 1988 1989 1990 1991 1992
Date
Figure 9.14 Hydrograph for Arapahoe Well A-6, Centennial Water and Sanitation
District, Highlands Ranch, Colorado.
Apparent
Net Flow per Hydraulic
Screened Screen Length Interval Conductivity
Interval (ft) (ft) (G/min) (ftlday)
m (400 ft) below land surface. The substantial water level fluctuations in
this well during cycle testing are shown in Table 9.14.
Sand was found in the recharge water as determined by a cartridge filter
and a Rossum sand tester. However, this was purged from the well during
recovery and no evidence of residual well plugging was found between
successive cycles. The sand was probably settled in the water distribution
system and was carried into the well during reverse flows occurring during
recharge. Sand content in the recharge water declined after the first cycle.
Upon completion of testing, it was recommended that the well should
be backflushed to waste at least every 4 weeks for a period of 4 to 8 hours
to maintain its capacity. It was also recommended that pH of the recharge
water should be maintained within a range of 7.5 to 8.3 to minimize
potential for precipitation of ferric hydroxide and calcium carbonate.
The ASR well is now in full operation and two additional wells are
being converted to ASR use. The objective of the ASR system is to meet
seasonal peak water demands through more efficient use of existing supply
and treatment facilities. All of the stored water will be recovered. Long-
term drought storage is also under consideration and is the subject of an
Alkalinity 92 123
Aluminum 0.05 0.06
Ammonia 0.43 0.17
Boron 0.05 0.10
Calcium 51 32
Chloride 28 3.25
Eh, millivolts (estimated) 400 200
Fluoride 0.95 0.75
Iron 0.015 0.13
Magnesium 10 1.85
Manganese 0.005 0.02
Nitrate 0.05 0.05
pH, laboratory 7.6 8.1
Phosphorus 0.02 0.005
Potassium 2.85 3.75
Silica 4.95 12.85
Sodium 29.5 27.5
Sulfate 91 16.5
Total dissolved solids 288 180
Note: All values expressed in milligrams per liter (mg/L) unless otherwise specified.
(a) To compute the average values, half the detection limit value was used when
the value for the parameter was below the analytical detection limits.
en
m
r
m
~
m
0
()
)>
en
m
en
-;
c
0
iii
TABLE 9.14 WATER LEVEL RESPONSE DURING ASR TESTING: HIGHLANDS RANCH, COLORADO en
Pump Test 300 24.2 - 24.5 (26.8 hr) 186.5 - 11 0.5 (26.8 hr)
Cycle 1 lnj 198 19.6 22.17 23.4 (121.6 hr) 186.5 241.5 272.9 (121.6 hr)
Cycle 1 Rec 309 25.5 32.9 34.0 (86.7 hr) 89.0 96.9 97.8 (86.7 hr)
Cycle 21nj 263 25.0 30.5 35.9 (364.8 hr) 195.0 263.7 381.8 (364.8 hr)
Cycle 2 Rec 410 39.0 45.5 55.9 (291.7 hr) 139.1 146.3 156.0 (291.7 hr)
Cycle 31nj 263 28.7 32.6 55.6 (1,125 hr) 174.9 225.7 467.5 (1,125 hr)
Cycle 3 Rec 410 36.6 40.6 40.4 (976.7 hr) 151.1 146.0 130.5 (976. 7 hr)
w
....
w
314 GROUNDWATER RECHARGE AND WELLS
Bask!, Inc
Denver. Colorado
( l-8CXJ-55-8ASKI)
r;c~~=·-lnjection
b!~~~ a~~~~._; Discharge
Las Vegas is located in the Mohave desert and receives about 100 mm (4
inches) of rainfall per year. Surrounding mountains may receive greater than
500 mm (20 inches) of precipitation, mostly in the form of snow. Water
supply to this area is supplied by the Las Vegas Valley Water District, which
operates what is currently (1993) the largest ASR system in the world with
a recharge capacity of about 300 megaliters/day (80 MG/day) and a recovery
capacity probably exceeding 380 megaliters/day (100 MG/day).
Recharge water is obtained from the treated water distribution system,
which is supplied from the Colorado River at Lake Mead. Water is low in
* Las Vegas Valley Water District, 3700 West Charleston Boulevard, Las Vegas, Nevada 89153
316 GROUNDWATER RECHARGE AND WELLS
250
~,Cemented gravel
72' Caliche
575
44' Cemented gravel
76 Cemented 0.75 in.
gravel 37'Caliche
3 r gravel 2 in.
6l'Callche
0.
0.
125' Cemented gravel
Lake Mead, both before and after treatment. Treated water is used for
recharge purposes.
The principal objective of this ASR system has been to seasonally store
available water to meet peak summer demands and also to meet long-term
water requirements. The system has been very successful in this regard,
having stored approximately 80 Mm 3 (21 BG, 65,000 acre ft) through
1993. Principal concerns have related to the quality of the recharge water
which, while meeting all standards, has a higher conductivity than the
native groundwater and a possible trend of increasing conductance in
recent years. Other water quality concerns have included the potential for
calcite precipitation in the aquifer, which has proven to not be a problem,
and the potential formation of trihalomethanes during storage. If THMs
form underground and are not reduced, they could possibly exceed ex-
pected future standards for drinking water. Investigations conducted by the
district have suggested that no THM reduction occurred during ASR
storage during the test period at this site.
The North Las Posas Groundwater Basin is about 29 km (18 miles) long
and 7 km (4.5 miles) wide, and is located in southern California near Los
* Calleguas Municipal Water District, 2100 Olsen Road, Thousand Oaks, California 91362
TABLE 9.16 RECHARGE WATER QUALITY: LAS VEGAS VALLEY WATER DISTRICT, NEVADA
Alkalinity (HCO) 159 161 171 159 154 155 159 157 159 154
Calcium 90 90 86 90 86 86 78 74 77 72
Magnesium 26 26 22 29 29 28 18 21 18 22
Chloride 90 92 104 108 90 90 53 55 64 64
Fluoride 0.37 0.39 1.00 1.00 0.31 0.28 0.28 0.28 0.29 0.29
Nitrate (N) 1.70 1.70 0.07 0.02 0.48 0.48 0.41 0.44 0.51 0.54
Potassium 3.90 4.30 6.0 6.0 3.7 3.7 4.2 4.2
Conductivity 1072 1057 1077 1035 1118 1107 846 857 829 833
(!lmhos em')
Sodium 105 107 109 134 115 115 70 71 63 69
Sulfate 290 290 311 309 301 301 232 216 211 206
pH (units) 8.20 8.20 8.20 7.80 7.95 7.75 7.92 7.84 7.76 7.53
Note: Analysis of Colorado River water in Lake Mead, before and after treatment.
Source: Brothers and Katzer (1 ).
320 GROUNDWATER RECHARGE AND WELLS
Dispersivity 22 ft (calculated)
Well efficiency 87%
Specific capacity 18 to 20 G/min/ft (initial production)
10 G/min/ft (injection, 1 hr)
7 G/min/ft (injection, 1 day)
4 G/min/ft (injection, 1 week)
2.5 G/min/ft (injection, 43 days)
Elevation Depth
(ftmsl) (ft)
Nested
Monitoring 250
Wells
I
750 ASRWell-97
Well-96
0
500
-250
250
-500
-500
-1500
-1250
-1500
500
Electrical Monitoring
lithologic Resistivity Well Completion ASR
Log (6' laterot log) Detoii Wall-97
Q~--~~._~.~----O~J=~lOO~~~(o~hm~~~)<c=~-~-TTr_~~-~----n--,,-,-Q
w T- !-- Cement--> · : Placement
100 -t---I~-~!----!J"'--i---I--'G£!rillo.,_t-----8·• oCG;out f--+-- 100
~ I MW-3 --+Y-++V' ~~~v;--"
200 ---+----!: : :!------H'-+--1--------!.VA t-0~f~own -- ! - I - 200
=~ 3oo-+--~:::!------4~~~~B~e~nto~n~rre~~
_ w >_ -->-v 1'61fets
~
§
2
<.!>
~
0
;il
.<:
1i
~
0
~
rzzzl Clay f.~,-~:-l Sand. Clay. and Gravel
[<<<J Sand f.!"--..) Sand and Gravel
c=:J Sand and Clay " Core Sample Location
Static water level was about 64 to 67 m (210 to 220ft) above mean sea
level or about 114 m (375 ft) below ground surface, which is at an
elevation of about 180 m (590 ft) above mean sea level. The well was
constructed in 1947. It has a 350 mm (14--inch) casing perforated from 204
to 283m (670 to 930ft) with slots measuring 5 mm (3116 inches) wide and
200 mm (8 inches) long. Figure 9.18 shows the well construction and also
lithology at this site. The well is equipped with a vertical turbine pump and
recharge occurs down the pump column after the pump impellers are
lowered against the bowl housing.
Recharge flow rates of up to 3.4 megaliters/day (620 G/min) were
achieved during the test program, while recovery rates were up to about
4.3 megaliters/day (800 G/min). Two ASR test cycles were conducted,
plus two step-drawdown pumping tests, one constant rate pumping test and
an injection test. The first test cycle was brief, to assess initial hydraulic
and water quality effects. The second test cycle injected 142 megaliters
(115 acre ft) over 43 days. This was then recovered; however, recovery
SELECTED CASE STUDIES 323
Calcium mg/L 25 23 68
Magnesium mg/L 16 15 13
Sodium mg/L 76 73 36
Potassium mg/L 4 3 3
Carbonate mg/L 2 1 0
Bicarbonate mg/L 99 100 189
Sulfate mg/L 500 53 46 121
Chloride mg/L 500 111 105 13
Total dissolved solids mg/L '1000 355 314 375
Hardness mg/L 200 128 118 222
Alkalinity mg/L 85 85 155
Color TON 15 10 10
Aluminum (unfiltered) 11g/L 1000 96 9
Aluminum (filtered) 11g/L 79 17
Arsenic 11g/L 50 1.80 0.00
Boron 11g/L 158 85
Cadmium 11g/L 10.00 0.00 0.00
Copper 11g/L 1000 0 0
Fluoride (total) mg/L 1.4-2.4 0.12 0.12 0.25
Iron (unfiltered) 11g/L 0 539
Iron (filtered) 11g/L 300 0 242
Lead 11g/L 50 0.00 0.00
Manganese (unfiltered) 11g/L 0 44
Manganese (filtered) 11g/L 50 0 43
Phosphorous 11g/L 0.04 0.03
Selenium 11g/L 10 0.00 0.00
Zinc 11g/L 5000 7.00 6.00
Ammonia (total): N mg/L 0.19 0.17
Nitrate (N0 3 ) mg/L 45 3.15 1.68 0.18
Silica mg/L 15.1 15.1 37.5
Total trihalomethane 11g/L 74 0
Chloroform 11g/L 4.2 0.0
Bromodichloromethane 11g/L 15.0 0.0
Dibromochloromethane 11g/L 31.0 0.0
Bromoform 11g/L 24.0 0.0
Total organic carbon mg/L 2.9 1.1
Total organic halogen 170 0
as Cl-
Field parameters
pH 8.40 8.25 7.52
Eh mv 419 211
Dissolved oxygen ppm 8.4 1.0
Temperature oc 18 16 25
Specific conductance 11mhos 500 600
SELECTED CASE STUDIES 325
able as a source of supply during 1956. This source supplies about 10,000
acre ft of water per year (9 MG/day, 34 megaliters/day) to the district.
Water from the reservoir is treated at the Corona del Mar water treatment
plant, which has a peak capacity of 136 megaliters/day (36 MG/day), and
utilizes coagulation, sedimentation, filtration, and chlorination processes
to treat the water before it is distributed to the service area.
Since 1963, the yield from the Cachuma reservoir has been supple-
mented by reactivated local wells that supply up to about 3400 acre ft per
year (3 MG/day, 11 megaliters/day) to Goleta Water District and some
other water users. In 1972 a "water shortage emergency" was declared due
to overdrafting of the aquifer. Some of these wells are equipped for both
production and recharge. Well depths range from about 300 to 1300 ft (91
to 396m).
The ASR storage zone is the same unconfined aquifer that originally
supplied the community with water. It is comprised of unconsolidated
clayey, silty sand alluvial sediments. Of the 28 wells used to recharge
treated drinking water during months when supply exceeds demand, 9 are
dual-purpose ASR wells while the remainder are single-purpose injection
wells. All were existing irrigation, private or public water supply wells,
retrofitted to meet ASR or injection needs. Recharge is typically down the
pump column, using both vertical turbine and submersible pumps.
Aquifer hydraulic characteristics are as follows:
Depth to static water level varies throughout the service area, reflecting
the terrain between the coast and the Santa Ynez mountains, and represent-
ing a principal constraint upon ASR system capacity. Minimum ground-
water levels are about 60ft below sea level. However, excessive recharge
volume would cause water levels to rise to land surface in portions of this
mostly confined system.
Native water quality in the aquifer varies from a TDS of 850 to 1000 mg/
L in the wellfield area, deteriorating toward the coast. Other constituents of
interest include localized areas of hydrogen sulfide at 0.5 to 1.0 mg/L, and
Fe plus Mn at about 1 mg/L. Recharge water quality is treated drinking water
from the treatment plant, which has a typical TDS concentration of 550 to
600 mg/L and a chlorine residual of 0.5 mg/L at the wellhead.
Recharge rates in individual wells range from less than 100 G/min (0.5
megaliters/day) to about 2.2 megaliters/day (400 G/min). Low average
326 GROUNDWATER RECHARGE AND WELLS
recharge rates of about 20 acre ft per year (12 G/min, 0.8 Lisee) per well
have been experienced in some wells. Recovery rates range from about
200 to 800 G/min (1.1 to 4.3 megaliters/day). The ratio of recharge to
recovery specific capacity is typically quite low, averaging about 30%. No
backflushing for redevelopment was conducted in any of the ASR wells
during the first 5 years of operation, relying upon extended pumping
during seasonal recovery to restore well specific capacity.
Since 1978, opportunities for recharge have been quite limited. Be-
tween 1978 and 1988, recharge occurred in only 7 years. Of these, all but
2 years had less than 448 acre ft per year (0.4 MG/day, 1.5 megaliters/day)
of recharge to the entire system. Maximum annual recharge during this
period was 1901 acre ft (1.7 MG/day, 6.4 megaliters/day).
The primary purpose of this system is long-term water storage, to
mitigate overdraft. All of the water stored is ultimately recovered, since the
basin is almost totally confined against lateral outflow due to the existence
of faults.
* City of Pasadena. Water Division, Water and Power Department, 150 South Los Robles Avenue,
Suite 200, Pasadena, California 9110 I
SELECTED CASE STUDIES 327
nities for storage. The huge potential savings derived from such regional
approaches to developing storage capacity should, in some cases, facilitate
enlightened and cost-effective water supply planning and implementation
for regional supplies.
Another element of the Pasadena ASR program is the consideration of
power generation during recharge. This was discussed under Section 3.2,
Design of ASR Wellhead Facilities.
above mean sea level. At Boulevard Park, static water level was 77 ft
below ground surface, at an elevation of 280 ft above mean sea level.
Water quality of the recharge water from the Cedar River pipeline and
the groundwater for the Riverton Heights production well are excellent, as
indicated by their low mean specific conductance values of 61 and 166
~-tmhos/cm, respectively. However, diatoms present in the recharge water
contribute to well plugging, which can be reversed by pumping.
The objective of the ASR test program was achieved, showing that the
Highline wellfield could be operated to meet seasonal peak demands
during summer months, while recharging the aquifer during winter months
to maintain water levels. Addition of a fourth ASR well to boost summer
recovery capacity is under consideration.
The test program included four ASR cycles conducted at an existing
unused test well adjacent to the Riverton Heights site. During the test
program, some well clogging was apparent. Detailed geochemical and
other investigations showed that the source of clogging was diatoms
(single-cell algae) present in the recharge water. Periodic backflushing of
the well with a frequency of about once every 2 weeks was sufficient to
prevent residual clogging and maintain recovery capacity.
To date, areawide water level response to seasonal wellfield operations
has been quite small. The opportunity exists at this site to utilize available
storage capacity more efficiently. Pumps could perhaps be set at the base
of the well screen rather than at the base of the casing, enabling an increase
in wellfield production rates and the associated interference between wells
during summer months. During the remainder of the year, wells would be
recharged to restore aquifer water levels to elevations at or above those
occurring prior to wellfield production. An increase in summer peak produc-
tion rates would be helpful to the Seattle Water Department, and would also
achieve operational benefits, since the effort required of operations per-
sonnel to operate the Highline wellfield is generally greater per unit of
water produced during the summer than that associated with the basic
supply of water from the Cedar River. Under pending regulations from the
EPA, filtration may be required for water from the Cedar River, in which
case the relative effort associated with the two sources would shift.
Another significant potential issue at this site is the pending regulation
pertaining to radon concentrations in drinking water. Depending upon
allowable radon concentration levels, the recovered water from the High line
wellfield may or may not meet the standards once they are promulgated.
Radon is not present in the recharge water but is picked up rapidly during
ASR storage. Treatment requirements would probably include aeration
and detention time in a ground storage reservoir which already exists at the
* Seattle Water Department, Water Management Department, 710 Second Avenue, Seattle, Washing-
ton 98104
330 GROUNDWATER RECHARGE AND WELLS
Riverton site but would need to be added at the Boulevard Park site.
Blending of the treated water with that from the Cedar River pipeline
should meet radon standards, if this becomes necessary.
9.17 KUWAIT*
During 1989, the Kuwait Institute for Scientific Research (KISR) initi-
ated an investigation of ASR feasibility to establish a strategic water
reserve. This work was performed for the Kuwait Ministry of Electricity
and Water. The goal was to store a large volume of drinking water in
brackish aquifers close to demand centers so that several months' supply
would be available, stored safely underground in the event of emergency
loss of seawater desalination plants. The water source for recharge would
be drinking water, which is primarily water from the desalination plants
blended with 5 to 10% brackish groundwater. Once established, this
strategic reserve would also be available to help meet summer peak
demands within the water supply service area. During the period October
1989 to May 1990, recharge investigations were conducted at 3 existing
well sites. The next phase of the program was interrupted by hostilities in
1990 to 1991, but has since continued.
Two sites were tested in the Dammam formation, a limestone artesian
aquifer, while one site was tested in the overlying Kuwait Group aquifer,
which is comprised of sand intervals between layers of cemented sandstone.
Both aquifers are brackish, with TDS levels ranging from 2700 to 5000 mg/
L in the Dammam formation and 3900 mg!L in the Kuwait Group aquifer.
Aquifer hydraulic characteristics were quite different at the three sites.
For Well SU-10 at Sulaibiya, the aquifer was composed of two produc-
ing intervals, at the top and bottom of the well, with the following hydrau-
lic characteristics:
The well construction included 124 m (406 ft) of 400 mm (16 inch)
casing, and open hole to 275 m (902ft). Of the many wells that have been
* Kuwait Institute for Scientific Research, PO Box 24885, Safat, 13109, Kuwait
SELECTED CASE STUDIES 331
tested for ASR feasibility, this is believed to be the one with the lowest
transmissivity. Static water level at the time of testing was at 37 m (120
ft) below land surface. A head difference of approximately 5 m (16 ft) is
believed to exist between the top and bottom producing intervals.
The well is equipped with a vertical turbine pump. A single test cycle
was conducted, replacing the pump with an injection tube to beneath the
static water level. A recharge volume of 16,416 em (4.3 MG) was injected
during a period of 30 days, at rates that varied from 0.65 megaliters/day
(121 G/min) initially to as low as 0.39 megaliters/day (73 G/min) in order
to avoid the water level rising above land surface. With redevelopment
during the injection period, a sustained injection flow rate of 0.58 megaliters/
day (109 G/min) could be maintained. Results indicated substantial plug-
ging, probably due to rust and sand in the distribution system that was
swept into the well at the beginning of recharge. When this was removed
during redevelopment by backflushing, hydraulic performance improved.
There was little evidence of geochemical plugging or air entrainment.
Following injection, the well was pumped at variable rates averaging about
0.98 megaliters/day ( 181 G/min). Pumping continued until background water
quality was reached. The specific capacity ratio between injection and
recovery was estimated at about 60%.
Background TDS in the aquifer was about 5000 mg/L. Mixing charac-
teristics were such that over 45% of the recharge volume was recovered
before the TDS exceeded 2000 mg/L in the recovered water, and 90% was
recovered before the TDS concentration reached 3000 mg/L.
Test results indicated that this site would be suitable for a strategic water
reserve, probably using only the lower producing interval of the Dammam
formation in order to reduce mixing between stored and native water. New
ASR wells would be required, with appropriate design and operation to
minimize clogging. Successive ASR cycles were expected to achieve
satisfactory recovery efficiency.
per year. Flowmeter logging indicated no head differences between the top
and bottom of the aquifer; however, about 60 to 75% of the flow occurred
at a depth of 372 to 375 m (1220 to 1230 ft). The water is brackish, with
a TDS concentration of about 2700 mg/L. Testing indicated the following
aquifer hydraulic characteristics:
This is a screen and gravel pack well with six 400 mm (16 inch)
screened intervals at depths ranging from 90 m (300ft) to 180m (600ft).
Screen intervals vary from 3 to 12m (10 to 40ft) in thickness and total43
m (140ft). The depth to static water level was 57.8 m (190ft) during the
test program. The vertical turbine pump was set between the second and
third screen intervals at a depth of about 122 m (400ft).
Hydraulic testing indicated the following aquifer and well characteristics:
was 13.7 G/min/ft. Injection was then resumed but was terminated after 22
hours when the rate of water level rise suggested that rapid plugging was
continuing. Total water stored was 4211 em ( 1.1 MG), less volumes
produced during the step test and pumping of spent acid. The well was then
placed back in routine operation and tracer concentrations were monitored
until 10,451 m 3 (2.8 MG) had been recovered at a typical production rate
of about 1.9 megaliters/day (0.5 MG/day).
Evaluation of potential causative factors suggested that air entrainment
and suspended solids in the recharge water probably caused the plugging.
Total suspended solids measurements yielded concentrations ranging from
3.2 to 7.4 mg/L during recharge. Substantial quantities of air and turbid
water were noted during redevelopment.
Mixing occurred in the well, partly due to differential plugging of the
six screen sections. Tracer testing was performed u,sing tritium as a tracer
during the latter portion of the injection test following acidization of the
well. It was also possible to use sulfate as a natural tracer. TDS concen-
tration of the native water was about 3900 mg/L while sulfate concentra-
tion was about 1050 mg/L. Injection water was obtained from the wellfield
collection system and had a TDS concentration of about 4400 mg/L but a
sulfate concentration of about 2000 mg/L. At 30% recovery, sulfate had
declined to 1910 mg/L; at 70%, it had dropped to 1650 mg/L, and at 100%
it had dropped to 1390 mg/L.
Although this test was conducted at a very small scale and encountered
a variety of complex logistic and technical difficulties, it is notable due to
the unique use of tritium as a tracer. Tritium was selected with the
assumption that natural tracers would not be sufficient; however, sulfate
differential concentrations in the recharge and native water also proved to
be useful. Both tritium and sulfate results indicated the same approximate
mixing curves during recovery.
Test results were not conclusive; however, they suggested the limited
potential value of this aquifer in the Sulaibiya area as a component of a
strategic water reserve utilizing both the Dammam and Kuwait Group
aquifers. Further effort would be required to design and operate ASR wells
in this aquifer in order to meet overall objectives while minimizing plug-
ging and meeting recovery water quality objectives.
Referenees
CHAPTER 1: INTRODUCTION
1. Postel, S., The Last Oasis: Facing Water Scarcity, The Worldwatch Environmental
Alert Series, Worldwatch Institute, W.W. Norton & Co., 1992, p. 35.
2. Postel, S., The Last Oasis: Facing Water Scarcity, The Worldwatch Environmental
Alert Series, Worldwatch Institute, W.W. Norton & Co., 1992, p. 30.
3. The World Almanac and Book of Facts, 1988.
4. Shiklomanov, I. A., Global Water Resources, Nature and Resources, Vol. 26, No.
3, 1990.
5. L'vovitch, M. 1., World Water Resources and Their Future, Washington, D.C.,
American Geophysical Union, 1979, as cited in van der Leeden eta!., The Water
Encyclopedia.
6. L' vovitch, M. I., Ground- Water Storage and Artificial Recharge, Natural Resources/
Water Series No. 2, United Nations, New York, 1975.
7. Limaye, S.D., Director, Ground Water Institute, Pune, India, personal communica-
tion, 1994.
8. Peters, J. H., Artificial Recharge and Water Supply in The Netherlands: State of the
Art and Future Trends, A. I. Johnson and R. D. G. Pyne, Eds., Proceedings of
American Society of Civil Engineers Second International Symposium on Artificial
Recharge of Ground Water, Orlando, FL, July 17-22, 1994.
9. Cederstrom, D. J., Artificial recharge of a brackish water well, Commonwealth,
Virginia Chamber of Commerce, Richmond, V. 14, No. 12, p. 31, 71-73, 1947.
10. Todd, D. K., Annotated Bibliography on Artificial Recharge of Ground Water
Through 1954, U.S. Geological Survey Water Supply Paper 1477, 1959.
11. Signor, D.C., Growitz, D.J., and Kam, W., Annotated Bibliography on Artificial
Recharge of Ground Water, 1955-67, U.S. Geological Survey Water Supply Paper
1990, 1970.
12. Knapp, G. L., Artificial Recharge of Groundwater. A Bibliography, Office of Water
Resources Research, Washington, D.C., 1973.
13. O'Hare, M.P., Fairchild, D. M., Hajali, P. A., and Canter, L. W.,Artificial Recharge
of Ground Water: Status and Potential in the Contiguous United States, Environ-
mental and Groundwater Institute, University of Oklahoma, Lewis Publishers,
Chelsea, Ml, 1986.
14. Merritt, M. L., Subsurface Storage of Freshwater in South Florida: A Digital Model
Analysis ofRecoverability, U.S. Geological Survey Water Supply Paper 2261, 1985.
15. Brown, D. L. and Silvey, W. D., Artificial Recharge to a Freshwater-Sensitive
Brackish-Water Sand Aquifer, Norfolk, Virginia, U.S. Geological Survey Water
Supply Paper 939, 1977.
335
336 GROUNDWATER RECHARGE AND WELLS
1. Caduto, M. J. and Bruchac, J., Keepers of the Earth, Fulcrum Inc., Golden, CO,
1988.
2. Merritt, M. L., Subsurface Storage of Freshwater in South Florida: A Digital Model
Analysis of Recoverability, U.S. Geological Survey Water Supply Paper 2261, 1985.
CHAPTER 5: GEOCHEMISTRY
1. Rice, G., Brinkman, J., and Muller, D., Reliability of Chemical Analyses of Water
Samples - The Experience of the UMTRA Project: Groundwater Monitoring
Review, 71-75, 1988.
2. Hem, J. D., Study and Interpretation of the Chemical Characteristics of Natural
Water, U.S. Geological Survey Water Supply Paper 2254, 1985.
3. Linn, J. C., Schnoor, J. L., and Glass, G. E., Sources and Fates of Aquatic Pollutants,
Advances in Chemistry Series, No. 216, American Chemical Society, Washington,
D.C., 209-228, 1987.
REFERENCES 337
I. CH2M HILL Inc., Environmental Risk and Geochemical Analysis Related to the
City of St. Petersburg's Underground Injection and Monitoring System, Engineering
Report for the City of St. Petersburg, FL, March 1993.
1. Brothers, K. and Katzer, T., Water Banking Through Artificial Recharge, Las Vegas
Valley, Clark County, Nevada, Journal ofHydrology, 115,77-103, Elsevier Science
Publishers B.V., 1990.
Appendiees
CONVERSION FACTORS
339
340 DRINKING WATER QUALITY STANDARDS
REGION IV
National Primary
Drinking Water Regulations
12/2211992
INORGANICS
Acrylarnide (1/91) TT 0
Alachlor (1/91) 0.002 0
Aldicarb (5/92) Deferred
Aldicarb sulfone (5/92) Deferred
Aldicarb sulfoxide (5/92) Deferred
Atrazine (1/91) 0.003 0.003
Benzene (7/87) 0.005 0
Benzo(a)pyrene (7/92) 0.0002 0
Carbofuran (1/91) 0.04 0.04
Carbon Tetrachloride (7/87) 0.005 0
Chlordane (1/91")· 0.002 0
2,4-D (1/91) 0.07 0.07
Dalapon (1 /92) 0.2 0.2
Dibromochloropropane (DBCP) (1/91) 0.0002 0
a-Dichlorobenzene (1/91,5/89) 0.6 0.6 0.01
p-Dichlorobenzene (7/87) 0.075 0.075
p-Dichlorobenzene (1/91,5/89) 0.005
1,2-Dichloroethane (7/87) 0.005 0
cis-1,2-Dichloroethylene (1/91) 0.07 0.07
trans-1,2-Dichloroethylene (1/91) 0.1 0.1
1,1-Dichloroethylene (7/87) 0.007 0.007
Dichloromethane
(Methylene chloride) (7/92) 0.005 0
1,2-Dichloropropane (1/91) 0.005 0
12/22/1992
MICROBIALS
RADIONUCLIDES
12/22/1992
FOOTNOTES
ANNEX I
USf OF PAltAMETERS
A. ORGANOLEmC PARAMETERS
Maximum
Exlft5$ion
Guide le>d admissible Comments
l'arameter.i of the (GL) oooceDtratioo
results(')
(MAQ
1
( ) If, on the basis of Directive 71/354/EEC as last amended, a Member State uses in its national legislation, adopted in
accordance with this Directive, units of measurement other than these indicated in this Annex, the values thus indicated
must have the same degree of precision.
Maximum
Expression
Guide level IIC!mUsible
Para mete~ of the Comments
(GL) concentration
results(;)
(MA(.)
5 Temperature "C 12 25
12 Magnesium Mg mgll 30 50
346 DRINKING WATER QUALITY STANDARDS
Maximum
Expression Guide level admissible
Parameten of the Comments
(Gl) concentration
results ( 1) (MAO
14 Potassium K mg/1 10 12
Expression Maximum
Guide level admissible
Parameten of the
(Gl) Comments
concentration
results(')
(MAC)
(') Certain of these substances may even be toxic when present in -.cry substantial q"""lities.
EUROPEAN COMMUNITY 347
Maximum
Expression Guide level admissible
Parameters of the (Gl) Comments
concentration
results (MAC)
23 Kjeklahl N mgfl I
Nitrogen
(excluding N
in NO, and
NO,)
28 Dissolved or II&II 10
emubified
hydrocarbons
(after
ex.traction
by petroleum
ether);
Mineral oils
34 Manpnese Mn Jlg/1 20 50
348 DRINKING WATER QUALITY STANDARDS
Maximum
ExpressiOn Guide level admi!sible
Parameters of the Comments
(GL) concentration
results (MAQ
39 Cobalt Co J.lg/1
Maximum
Expression Guide level admissible
Parameters of the Comments
(GL) coocentration
results(') (MAQ
Maximum
Expression Guide level admissible Comments
Parameters of the (GL) conc:entration
results (MAQ
44 Arsenic As ~gil 50
45 Beryllium Be ~gil
46 Cadmium Cd ~g/1 5
47 Cyanides CN ~gil 50
48 Chromium Cr ~gil 50
49 Mercury Hg ~gil I
50 Nickel Ni ~gil 50
52 Antimony Sb JJg/1 10
53 Selenium Se~tg/1 10
----
54 Vanadium v jlg/1
56 Polycyclic
aromatic
~gil I 0·2 -reference substances.
-tluoranthene
hydrocarbons -benzo-3, 4-0uoranthene
-henzo-11, 12-Huoranthene
-benzo-3, 4-pyrene
--benzo- I, 12-perylene
-indeno (I ,2,3-cd) pyrene
350 DRINKING WATER QUALITY STANDARDS
E. MICROBIOLOGICAL PARAMETERS
60 Sulphite-ocducing
Clostridia 20 - - MPN.:s:l
Water inteuded for human consumption should not contain patlJosenic organisms.
If it is necessary to supplement !be microbiological analysis intended for human consumption, the samples should be
examined not only for the bacteria referocd to in Table E but also for pathogens including:
- salmondla,
- patllqpenic staphylococci.
=
- feal bacteriopbqes,
- entero-viruoes;
not should such water contain:
=
-parasites,
Results: Maximum
size of Guide level admissible
Parameters (GL) Comments
sample concenttation
(ioml) (MAC)
(')For disinfected water the corresponding values should be considerably lower at the point where it leaves the prooessing plant.
(')If, during succesive sampling, any of these values is consistently exceeded a cbeck should be carried ouL
EUROPEAN COMMUNITY 351
F. MINIMUM REQUIRED CONCENTRATION FOR SOFTENED WATER INTENDED FOR HUMAN COMSUMPTION
)""~~·-= . -~
2 Hydrogen ion concentratioo pH
4 Dissolved oxygen
N B:- The provisions for hardness, hydrogen ion oonoentration, dissolved oxygen and calcium also apply to desalinated water.
-If, owing to its excessive natural hardness, the water is softened in accordance with Table F before being supplied for
consumption, its sodium content may, in exceptional cases. be higher than the \'lllues given in the 'Maximum
admissible conoentration' column. However, an effort must be made to keep the sodium content at as low a level as
possible and the essential mquirements for the pmleetion of public health may not be disregarded.
ANNEX II
............
-
Swtdard ma.lym moaisoriac
....
Oa:alioaal
,_......, M1n1mum moaitonna Current monitorifts Pcriocbc~ orile8!1toi
........... (C I) (C 2) iC )) (C 4)
c UNDESIRABLE -nitrates
PARAMETERS -nitrites
-ammonia
D TOXIC
PARAMETERS
Now An iniual analysis. to be earned out before a source is exploited, should be added. The parameter.< to be considered
would be the current monitonng analyses plus intn alia various toxic or undesirable substances presumed present. 1be
list would be drawn up by the competent national •uthorities.
( 1) Qualitative assessment.
f) Except for water suppied m contame~.
(') Or other disinfectanu and only in the case of treatment.
(') These parameters will be determined by the competent national authority. laking account of all factors which might affect
the quality of drinki"' water supplied to users and which could enable the ionic balaDce of the constituents 10 be assaaed.
(') The competent national authority may usc pammeters other than thooe mentioned in Allllell Ito Ibis Dinocli"Ve.
EUROPEAN COMMUNITY 353
---
(') Frequency left 10 lbe cliscmion of the competent national authorities_ H~r. water intended for the food-
IIWIUfocturina industries must be monitored at least ooce a year_
(')The competellt beallh authorities should cudeavour to inause this frequency as far as their resources allow_
(') (a) In the case of woter which must be disinfected, miaobiological analysis should be twice as f!"'4uent_
(b) Where analyses are >'Cf)' frequent, it is advisable to take samples at lbe most rqular intervals possible_
(c) Whore lbe values of the resuhs obuincd hom aamples taken durin& the preceding years are constant and significantly
be1t<r than lbe limits laid clown io Anoex l, and where oo factor likely 10 cause a deterioration in the quality of
the water has been dioc:overcd, the minimum frequencies of the analyses refencd to above may be reduced:
- for surface woters. by a factor of 2 with the exc:epioo of the frequencies laid down for microbiological analyses:
- for pound woters, by a l'actor of 4, but without prejudice to the provisions of point (o) above_
354 DRINKING WATER QUALITY STANDARDS
Annex 2
a Immediate 1nvest1gat1ve act1on must be taken if either E. coil or total coliform bactena are detected
The m1n1mum act1on m the case of total coliform bactena 1s repeat sampling; if these bactena
are detected in the repeat sample, the cause must be determmed by 1mmed1ate further
1nvest1gauon.
b Although E. coli is the more prec1se ind1cator of faecal pollution, the count of thermotolerant
coliform bactena IS an acceptable alternative. If necessary, proper confirmatory tests must be
earned out. Total coliform bactena are not acceptable md1cators of the sanitary oual1ty of rural
water supplies, particularly 1n trOPICal areas where many bactena of no san1tary s1gnif1cance
occur 1n almost all untreated suppl1es
c It 1s recogmzed that, in the great ma]onty of rural water supplies m develoomg countnes, fae-
cal contamination IS Widespread. Under these cond1t1ons, the nat1onal surveillance agency should
set medium-term targets for the progress1ve Improvement of water supplies, as recommended
1n Volume 3 of Guidelmes for dnnking-water quality.
356 DRINKING WATER QUALITY STANDARDS
selen1um 0.01
uranium NAD
WORLD HEALTH ORGANIZATION 357
B. Organic constituents
Chlormated alkanes
caroon teuachlonde 2
d1cniorometnane 20
1, 1-dJchJoroetnane NAD
1,2-dJchloroetnane 30~ ior excess nsk of 10-:
1,1 ,1-tnchloroethane 2000 (PI
Chlormated ernenes
v1nyl chloride 5D for excess nsk of 10- ~
1,1-dJchJoroetnene 30
1,2-dJchloroethene 50
tnchloroethene 70 (PI
tetrachloroethene 40
AromatiC hydrocarbons
benzene lOb for excess r1sk of 10-:;
toluene 700 ATO
xy1enes 500 ATO
ethyl benzene 300 ATO
styrene 20 ATO
benzo[a]pyrene 0 7° for excess nsk of 10-:
Ch/onnated benzenes
monochlorobenzene 300 ATO
1,2-dJchlorobenzene 1000 ATO
1,3-dJchlorobenzene NAD
1A-dichlorobenzene 300 ATO
trichlorobenzenes (total) 20 ATO
Miscellaneous
di(2-ethylhexvlladlpate 80
di(2-ethylnexyllphthalate 8
acrylam1de 0.5b tor excess nsk of 10- 5
ep1chlorohydnn 0.4 (Pi
hexachlorobutadlene 0.6
edet1c acid (EDTAI 200 (PI
nJtrilotnacetJc ac1d 200
dJalkyltms NAD
tnbutyltJn ox1de 2
358 DRINKING WATER QUALITY STANDARDS
C. Pesticides
monochioram1ne
01- ana tncn1oram1ne
cn1orme 5 ATO Fer eftect1ve O!Sinrect!On tner-=
snouio De a res1aual concentration c'
iree cn1orme of ~0 5 mg;ime aner a~
least 30 m1nutes co mac: t1me at c"
<8.0
ChlOrine diOXIde A gu1oe11ne value nas not oeen estac·-
llsnec oecause oi me raDIO oreakOO\i\T
of cn1onne orox1oe ana oecause tne
cn1ome gu1de11ne value IS aaequate1 '-
orotecnve ior ootent1a1 tox1c:tv tron-
cnlonne OIOXIOe
IOdine NAD
Disinfectant Guideline value Remarks
by-products (J.Lg/litre)
haiogenareo aceronJtriles
dlcnioroacetonJtrlle 90 !PI
dioromoacetonmlie 100 IPI
oromocnloroacetonltw:· NAu
wen 1oroaceto nnr11e 1 IP
cvanogen cn1onae 7G
1as CNI
cniorop1cnn NAD
a IPI - Provtstonal QU!Oellne value Thts term ts usee ror constttuents ror whtch mere 1s some
ev1aence 01 a ootenttal nazare o•;t wnere tne avatiaote tnrormat1on on neattn efrec;s 1s nmne:
or wnere af"' uncenatntv iactor greater man 1000 nas oeen usee 1n me aer1vat1on or me lO'e•~
able Jailv IntaKe lTD I I Provtstonal gutdeilne va1ues are a1so recommenced 111 ror suostancec
ror whtcn the ca1culatea gu1denne va1ue wouia oe oelow tne oracucal auanttitcauon 1eve1. o·
oelow the tevel tnat canoe acntevea tnrougn oracttcal treatment metnoas: or 12t wnere OISir'-
fectton 1s itkelv to result tn me (JUIOellne value betng exceeaec
0 For substances that are constaerec to oe carc:nogen1c, the gutdeitne va_tue ts :ne concentrattor
1n arlnKtng-water assoc1ateo wnn an excess iifettme cancer nsk of ~~-o tone acantonal cance·
oer 100 000 ot me popuJal!on tngesttng artnKtng-water comatntng tne suostance at :ne gutoe~
l1ne value for 70 vearsl. Concentrations assoctated wnh esnmateo excess lifetime cancer nsKs
of 10-~ and 10- 6 canoe calculated ov muluoivtng and dtvtdtng. resoecttvely, me gutoe11ne
va1ue by 10. _
in cases 1n whtch the concentration associated wtth an excess liiellme cancer nsk of 10-:
ts not teas101e as a result of tnaoequate anaivllcal or treatment tecnno1ogy, a orovts1onai gutae-
itne va1ue ts recommenoeo at a practicable levet ana the esnma~ea assoctated excess liteume
cancer nsk presented~
It snould be emonastzed that tne guideline values for carc1nogentc suostances nave oeen
computed from nypotheucal matnemattcai models that cannot oe venfted exoenmentallv ana
that me va1ues snouid oe tnteroreted differently than TDI-oasea values oecause ot the lacf:
of orec:s1on of me models~ At oest. these values mustoe regardea as rougn esnmates of cance:
nsk~ However. tne models usee are conservative ana orooablv err on the stde of cautton. Macerate
snon-term exoosure to 1eve1s exceeamg tne gu1aeime value for carctnogens aces not stgn1i 1-
cantlv affect tne nsr-
c NAD- No aaeauate aata to oerm!l recommendation oi a healtn-oasea gutdel!ne value
0
ATO- Concemranons of tne suostance at or oe1ow the health-cased gutdeline value mav a'-
fect the apoearance, taste. or oaour ot the water
WORLD HEALTH ORGANIZATION 361
Chemical Remarks
asoestos
Sliver L'
ttr·
; .
r
gross aiona actiVIt'( 1
\...'I li a screenmg vaiue IS exceeaec. more
gross oeta act1v1ty deta!leo rad1onuc!1de analYSIS 1s neces-
sary f-'1or,er Yaiues ao not necessarw.
1mp1v mat me water IS unsunao1e ior
human consumot1on
362 DRINKING WATER QUALITY STANDARDS
Prws1ca1 oaramerers
COlOUr 15 TCU 0 aooearance
taste ana oaou· snou1o oe acceotaoie
temoeraturs snoulo oe acceotaole
turDIOill' 5 NTU 0 aooearance; for efiect1ve te:mma1 01S·
mrecnon, meo1an curb1dttV ::;:lNTL,
s;ngle samo1e ::;: 5NTU
1norgamc consuruenrs
aiummtum 0.2 mg/1 OeOOSiliOnS, OlSCOIOratiOr'
ammon;a i.:i mg1l ooour ana taste
chlortae 25C> mg/1 tasTe, corrostor:
co poe: 1 mg;l srammg of launorv ana san1tarv ware
:r:ea!m-oasec orovts1onat gu,oe11nco
va1ue 2 mg/i!Wo
nard ness n19h nardness: scale deoosttton, scurr
format ton
10w naraness: ooss1bie corrostor
hvdrogen suifiae O.C6 mg/1 oaour and taste
~ron 0.3 mgil staintng of iaunarv and santtary ware
manganese 0.1 mg/1 stammg of launory and santtarv ware
tnealth-oasea proviSIOnal gUtaellne
value 0.5 mg/lmel
diSSOlved oxygen matrect effects
oH lOW pH: corroSIOn
n1gn oH: taste, soaov fee'
oreferabiv < 8.0 for eftect1ve dismtec-
tton wltn chtorme
sod1um 200 mg/1 raste
sulfate 250 mg/1 taste, corros1on
total atssolvea soltas 1000 mg/1 taste
ZinC 3 mg;l aooearance. taste
Orgamc consmuenrs
toluene 2.1-170 ,ug/1 oaour, taste lnealth-baseo gu1deltne
value 700 llQ/1)
xylene 20-1800 ,ug/1 odour, taste mealth-baseo gutdeline
value 500 ,ug/1!
ethvlbenzene 2-200 ,ug/1 odour, taste !health-basea guideline
value 300 ,ug/11
stvrene 4-2600 ,ug/1 oaour, taste lhealth-oasea gu1deitne
vaiue 20 ,ugili
WORLD HEALTH ORGANIZATION 363
a The levels tndtcatea are not orec1se numoers Proolems mav occur at tower or ntgner values
accord1ng TO local ctrcumstances. A range or taste and oaour thresnola concemrat1ons tS g1ver.
tor organ1c constituents
0 TCU. ume colour unt~.
CONVERSION FACTORS
Multiply By To obtain
Metric Unit Ene,lish Unit
Length
kilometre (km) 0.6214 mile
metre (m) 1.0936 yard
centimetre (em) 0.0328 foot
millimetre (mm) 0.03937 inch
Area
sq kilometre (km 2 ) 0.3861 square mile
hectare (ha) 2.471 acre
sq metre (m 2 ) 10.764 square foot
sq metre (m2 ) 1550 square inch
sq centrimetre (cm 2 ) 0.1550 square inch
Volume
cu centimetre (em') 0.061 cubic inch
cu metre (m') 1.308 cubic yard
litre (L) 61.02 cubic inch
litre (L) 0.001308 cubic yard
litre (L) 0.2642 US gallon
litre (L) o.:n Imperial gallon
Wei~t
metric tonne (t) 0.984 long ton
metric tonne (t) 1.102 short ton
kilogram (kg) 2.205 pound,avdp
gram (g or gr) 0.0353 ounce, avdp
Other
cu centimetre (em') 0.0338 fluid ounce
kilograms/sq em (kgicm') 14.225 pounds/sq in
metric horsepower (CV) 0.9863 hp
kilowatt (KW) 1.341 hp
bar 14.5 psi
Flow
megaliters/day (mlld) 0.264 million gallons/day (MGD)
Temperature
·c = 519 CF- 32)
Viscosity of Water
Temperature ( 0 F) 32 50 60 70 80 100 120
Temperature (•C) 0 10 15.6 21.1 26.7 37.8 48.9
Kin. Vise. (CS) 1.79 1.31 1.12 .98 .86 .69 .57
CONVERSION FACTORS 365
CONVERSION FACTORS
Multiply By To obtain
EDJ;Iish Unit Metric Unit
L~th
mile st.arute (m) 1.609 kilometre
yard (yd) 0.9144 metre
foot (ft) 0.3048 metre
inch (in) 25.4 millimetre
Area
sq mile (mile2 ) 2.590 sq kilometre
acre 0.4047 heclllre
sq foot (ft 2 ) 0.0929 sq metre
sq inch (in2 ) 0.000645 sq metre
Volume
cu yard (yd') 0.7645 cu metre
cu inch (in 3 ) 16.387 cu centimetre
cu foot (ft 3 ) 0.0283 cu metre
cu inch (in 3 ) 0.0164 litre
cubic yard (yd 3 ) 764.55 litre
US gallon (US Gal) 3.785 litre
US gallon 0.833 Imperial gallon
Wei!;ht
long ton (l g ton) 1.016 metric ton
short ton (sh ton) 0.907 metric ton
pound (!b) 0.4536 kilogram
ounce (oz) 28.35 gram
Other
fluid oz (fl oz) 29.57 cu centimetre
pounds/sq in 0.0703 kilogram/sq em
psi 0.0689 bar
horsepower (hp) 1.014 metric horsepower
horsepower (hp) 0.7457 kilowatt
F1ow
million gallons/day 3.785 megaliters/day (Ml/d)
Tem~rature
OF= 915 ("C+32)
Index
A applications, alternative; ASR non-
technical issues, selected; ASR
Acidification, 4 program development; ASR systems,
Acidization, 135, 301 design of; ASR technical issues,
Aeration, 158, 216 selected
Aggressive water, 20, 26 backflushing operations, 65
Agriculture construction details, 72
runoff, nutrient reduction in, 19, 26 cycles, 48, 160
water demands for, 246 definition of, 6-8
water supply, 18, 26 expansion, 218
Air bubbles, 81, 113 facilities
Air-conditioning return flows, I 0 supply-demand relationship with, 31
Air lines, 88 unit costs for, 218, 219
Alkalinity, 51, 171, 176, 296 feasibility assessment, 25, 110
Alternative pump setting, 74 implementation, 154
Alum coagulation, 265 operations, cost of, 40
American Water Works Association projects, planning of, 23
(AWWA), 68 recovery
Ammonia, 152, 181 duration, 44
Ammonium, 190 efficiency, 106, 107
Ammonium acetate, 215 site
Anion chemistry, 179 investigations, 216
Annulus recharge, 81 hydrogeology, 273
Artificial recharge, 4 storage, 109, 145, 149, 224
Aqueduct system, 320 arsenic reduction during, 259
Aquifer(s), 32 nutrient reduction during, 259
bacteria present in, 200 stacking, 58
basalt, 2, 233, 237 zone, 70, 211
characteristics, 209 success, 87
freshwater, 237 test, 46, 50, 108, 292
glacial drift, I 0 as water management alternative, 238
hydraulics, 305, 330, 332 well(s), 135
limestone, I 00 construction, 294
lithology of, 32 design, I 05, 254
mineralogy, 117 location of, 58
parameters, 119 recharge, 7
pH stabilization in, 136 testing characteristics, 124
plugging, 102 well field
pretreatment, 154 designs, alternative, 99
recovery efficiency in, 108 site, native water at, 97
sand, 13 wellhead
storage, 143, 217, 224 design, 75
environmental benefits of, 237 filtration, 139
recovery, 15 Aquitards, 32
transmissivity, 107 Arsenic, 157, 324
unconsolidated, 135 Artesian aquifer, 309
water quality of, 32 Artificial recharge, 3, 4, 9, II
Aquifer storage recovery (ASR), 4---6, see also ASR, historical development of, 9-17
ASR, historical development of; ASR Australia, 13-17
367
368 GROUNDWATER RECHARGE AND WELLS
D
Dams, 3
E
Data collection system, computer-based, 62 Ecosystems, impacted, 39
DBP, see Disinfection byproduct Eh, see Hydrogen electrode
Delivery points, 263 Electric motor damage, 137
Denitrification, 182 Electrical breakdowns, 52
Density stratification, 106 Electrical outlets, 95
Deoxygenation, 153, 216 Electronic pressure transducers, 88
INDEX 371
Emergency Geochemical
demands, 223 analysis, 33, 171, 298
recovery, 59 assessment, 156
storage, 17, 26 measurements, 169
Equilibrium modeling, 159
calculations, 208 plugging, 47, 21 I
conditions, 186 simulation, 37
Erosion, rates of, 2 Geochemistry, 169-216
Evapotranspiration, 107 aquifer characteristics, 191-198
colored pictures of cores, 193-194
mineralogy, 194
F
physical characteristics, 193
Feasibility investigations, 237 scanning electron microscopy, 194-196
Feldspars, 196 thin section petrography, 196-198
Ferric hydroxide, 152, 153, 156, 312 field investigations, 216
Ferrous carbonate, 150, 153 geochemical models, 208-211
Ferrous sulfide, 150, 152 EQ3NRIEQ6, 210-211
Field data collection, 52 MINTEQ, 210
Field investigations, 24 PHREEQE, 210
Filter backwash operations, 304 WATEQ, 209-210
Filtration, 260, 325 geochemical processes, 199-208
Fiscal questions, 221 adsorption, 203
Fish hatchery temperature control, 20, 26 biofouling, 200-203
Flood control, 3 dissolution, 204--207
Flooding, 65 ion exchange, 203-204
Flow kinetics, 208
distribution, 165 oxidation, 204
sequence, 202 suspended solids clogging, 199-200
Flowmeter(s), 89 laboratory testing, 211-216
failure, 52 batch testing, 215-216
logs, 42, 332 column testing, 211-215
range, 89 water chemistry, 171-190
Fluoridation, 328 changes in groundwater chemistry with
Flushing period, 76 depth, 183-186
Flux, 120, 124 Eh-pH diagrams, I 86-190
Food grade oil, 85 hydrogeochemical processes, 180-183
Freshwater mass balance, 177-178
maintenance of, 77 parameters, 171-177
recovery of, 159 water chemistry diagrams, 178-180
zones, 4 Geophysical logging, 42, 73
Friction loss, 83 Giardia Iamblia, 225
Future directions, 257-263 Glacial drift aquifer, 10
global applications of ASR, 261-263 Glauconite, 196
constraints, 262 Gravel pack, 73, 114
driving forces, 26 I -262 Groundwater
opportunities, 262-263 chemistry, 180, 184, 205
regulatory issues, 260-261 iron present in, 151
technical developments, 258-260 levels, restoration of, 18, 26
management, 227
production, 320
recharge, 6
G systems, dissolved iron in, 186
Gate valve, 142 untreated, 241, 248
Gel filtration, I 15 velocity, 8
372 GROUNDWATER RECHARGE AND WELLS
withdrawals, 2 Karst
Gypsum, 176 limestones, 244
solution features, 61
Kinetics, 208
H
HAAs, see Haloacetic acids
Halloysite, 196
L
Haloacetic acids (HAAs), 142, 148 Laboratory batch testing, 170
Homeowner opposition, 65 Leaching, 215
Horizontal wells, 245 Levees, 3
Hurricanes, I 06, 275, 281 Limestone aquifer, I 00
Hydraulic conductivity, 131, 212, 327 Lithology, 306, 316
Hydraulic data collection, 52 Loading rate, 120
Hydraulic testing, 306
Hydrogen electrode (Eh), 172
Hydrogen sulfide, 160
M
Hydrogeology, 31, 163 Magnetic flowmeters, 91
evaluation, 32, 34 Manganese
modeling, 37 concentrations, cyclic improvement in, !58
Hydrologic analysis, type of, 27 control, 155
Hydroxyl ion concentration, 174 dissolved, 188
organic, 215
oxidation, 185
reactions, 156
Idle capacity, 31 sources, 51
Illite, 196 stability and equilibrium, 189
Industrial cooling, 249 Marcasite, 204
Industry, water demands for, 246 Mass balance, 177, 268
Initial cycle recovery, 105, 109 Mechanical integrity testing, 80
Injection, 333 Membrane filter
rates, range of, 79 index (MFI), 122, 1,23
tubes, 69, 84, 141 measurements, 127
well(s), 4, 6 stages of clogging on, 114
for potable aquifer recharge, I 0 technology, 139
plugging of, 235 Methane-generating bacteria, 173
salinity intrusion barriers, 249 MFI, see Membrane filter index
single-purpose, 5 Microfiltration systems, 138
treatment of with clay stabilizer, !59 Mineral
Institutional constraints, 39 solubilities, 205
Intake structure, 62, 161 structure, 205
Ion exchange, 179, 212, 253 Montmorillonite, 196
Iron Multiple zones, 58
bacteria, 243 Municipal water systems, 13, 31
carbonate, 185, 204
concentrations, cyclic improvements in, !54
hydroxide, 203, 207, 215
N
Irrigation National emergencies, 260
needs, agricultural, 290 Native water quality, 267, 280, 296, 312, 325
season, peak, 250 Natural disinfection, 227
Irrigation, 247, 249 Natural environments, approximate position
of, 187
Natural iron, 203
K Negative pressures, 78, 141
Kaolinite, 196, 277 Nephelometric turbidity units (NTU), 122, 251
INDEX 373
manganese present in, !50 concentration, 46, 48, 61, 228, 290, 333
permeability, 93 background, 331
seawater in, 277 Total organic carbon (TOC), 143, 251, 296
selection, 73, 242 concentrations, 233
stacking of, 33 reduction, 252
transmissivity of, 105 Total suspended solids (TSS), 51, 112, 171,
Stored water 280, 296
advective loss of, 106 concentration, 243
bubble of, 56 measurement, 333
movement of, 97, 165 Tracer constituent, natural, 52
recovery of, 6 Transmissivity, 242, 266, 290, 299, 306, 332
stacking of, 266 Treatment
Storm drainage system, 69 alternatives, 261
Straddle packer tests, 290 plant, 62
Strategic water reserve, 26, 59 Trenching machine, 245
Stratigraphy, 32 Trickle flow, 42, 61, 77, 116
Streamflow diversions, 18, 26 Trihalomethanes (THMs), 18, 296
Submersible pumps, 82, 85, 86, 94, 276, 307 concentrations, 143, 144
Subsidence, 18, 26 reduction, 318
Sulfate-reducing bacteria, 173, 203 total, 324
Sulfide mineralization, 183 Trilinear diagram, 178, 181, 182
Sulfonator, 299 Tritium, 333
Sulfur TSS, see Total suspended solids
Eh vs. pH, 190 Turbidity, 171, 229, 230, 280
species, 189 Turbine efficiency, 96
Surface basins, 13
Surface recharge, 221
expansion areas, 15 u
projects, 9 UIC, see Underground injection control
systems, 4 Ultrasonic flowmeters, 91
Surface salinity barrier leakage losses, 19 Unconsolidated aquifers, 135
Surface storage reservoirs, 217 Underground injection control (UIC), 39, 227,
Surface Water Treatment Rule (SWTR), 225 246,261
Suspended solids, 114, 123, 138, 176 United States Geological Survey (USGS), 13,
content, 29, 132, 242 14
measurements, 121 Urban areas, ASR facilities in, 35
SWTR, see Surface Water Treatment Rule USGS, see United States Geological Survey
Synthetic streamflow, 162
System
capacity, 224 v
operational test, 27 4 Vacuum
reliability, 268, 326 breaker valves, 80
degassification, ISS
T Vegetation, transpiration from desert, 9
Velocity, 124
Target storage volume, 274, 285 Vermiculite, 196
TDS, see Total dissolved solids Vertical turbine pumps, 84, 94, 267, 283
Thermodynamic computer models, 208 Video log, 36
Thin section petrography, 192, 196
THMs, see Trihalomethanes
TOC, see Total organic carbon w
Total dissolved solids (TDS), 51, 172, 230, Wastewater
280 cost of treating, 2S3
376 GROUNDWATER RECHARGE AND WELLS